Note: Descriptions are shown in the official language in which they were submitted.
~2~7~3
LARGE PORE CATA~YSTS FOR HEAVY HYDROCARBON CONVERSION
Technical Field
This inven tion relates to an improved catalyst, a method for
its preparation, and a process for its use in the conversion of
5 petroleum oil feeds containing hydrocarbon molecules of high
molecular weight ~heavy). More particularly, the invention is
related to a catalyst composition comprising a catalytically ac tive
crystalline aluminosilicate zeolite uniformly dispersed within a matrix
component having large feeder pores for conveying reactants to and
10 reaction products from the zeoli-tic component.
Back~round of the Invention
In general, gasoline and other liquid hydrocarbon fuels boil in
the range of about 38C to 343C (100F to 650F). However, the
crude oils from which these fuels are made contain mixtures of
15 hydrocarbons which boil over wider temperature ranges, -the boiling
point of each hydrocarbon depending upon its molecular weight. As
an alternative to discarding or otherwise using the higher boiling
hydrocarbons, the petroleum refining industry has developed a
variety of processes for breaking or cracking the large molecules of
20 high molecular weight into smaller molecules which boil wi thin the
above boiling range for hydrocarbon fuels. The cracking process
which is most widely used for this purpose at the present time is
known as fluid catalytic cracking (FCC) and may employ a fluidized
bed reactor with backmixing and/or a riser reactor with progressive
25 flow. In a typical FCC process, feedstock oil is mixed with
particulate catalyst at an elevated temperature in the lower portion
of an elongated reaction vessel called a "riser". Contact of the hot
catalyst with the oil rapidly generates large volumes of gases which
propel the stream of feed and catalyst as a suspension through the
30 reaction zone at high velocity, giving relatively short contact time.
~k
RI-4078C '~
_`r~
~Z~47139
--2--
The initial propelling gases are comprised of vaporized oil, the
major portion of which boils below 552C (1025F) and is immediately
vaporized by contact with the hot catalyst which enters the riser at
a higher -tempera ture . A~ the suspension travels up the riser, a
5 large fraction of the feedstock hydrocarbons is converted to lower
boiling hy~drocarbons by catalytic cracking and these cracked
products form part of the propelling gases. The velocity of the
suspension is sometimes increased further by introducing diluent
ma-terials into the riser ei-ther along with the feed or separately.
10 The conversion reaction initiated in the lower portion of the riser
continues until the catalyst and gases are separated, which may
take place as the suspension leaves the riser reaction zone or in an
upper, larger diameter vessel for collec-ting the catalyst. Upon
being separated from the catalyst, the gases are usually referred to
15 as "product vapors".
Crude oil in its natural sta-te contains a variety of materials
which, unless removed prior to the cracking reaction, tend to have
troublesome effects on FCC processes. These include coke
precursors, such as asphaltenes, polynuclear aromatics and high
20 boiling nitrogen containing molecules; and metals, such as sodium
and small amounts of other alkali or alkaline earth metals, nickel,
vanadium, iron and copper, which are detrimental to the conversion
process and/or to the catalyst.
During the cracking operation, coke precursors either tend to
25 deposit as solid aromatic structures having some residual hydrogen
or are high boiling and do not vaporize but lay down on the
catalyst as a liquid. These coke deposits block the catalytically
active acid sites of the catalyst and thereby reduce its conversion
activity. While it is believed that both the solid and liquid
30 components of coke may cover and thereby block acidic sites, the
liquid components may also fill pores of the matrix and thereby
RI-4078C
~Z4~7~
--3--
retard diffusion of lower boiling components to the ~eolite.
Although the carbonaceous material formed by the conversion
process is referred to as coke, it may have hydrogen to carbon
ratios of 1. 0 or greater and may contain in addition to hydrogen
5 various amounts of other element depending upon the composition of
the feed. The coke formed is deposited on the catalyst particles
and thereby reduces the conversion activity of the catalyst. In
order to res-tore conversion activity, the contaminated catalyst is
regenerated by burning off the coke by contacting the catalyst
10 particles at relatively high temperatures with an oxidizing gas such
as air. The regenerated catalyst may -then be returned to the
reaction zone for additional passes or conversion cycles in contact
with fresh feed.
In general, the coke-forming tendency or coke precursor
15 content of a feedstock oil can be ascertained by determining the
weight percent of residue remaining upon pyrolyzing a sample of
the feed. Two -tests presently recogni2ed by the industry are the
Conradson carbon residue test described in ASTM D189-76 and the
Ramsbottom carbon test described in ASTM D524-76. In
20 conventional FCC practice, Conradson carbon residues of about 0.05
to 1.0 are regarded as indicative of relatively contaminate free gas
oil feeds.
Unless removed by careful desalting of the crude oil, the
sodium, and other alkali or alkaline earth metals can diffuse to the
25 active, i . e ., acidic, sites of the catalyst and poison of kill their
catalytic activity. Vanadium, and to a lesser extent nickel and
other metals, may also migrate to and poison acidic si tes . There
metals will therefore be referred to collectively as poison metals.
Nickel, vanadium, copper and iron are also known as "heavy
30 metals" and catalyze undesirable dehydrogenation reactions which
increase the amount of coke deposits on the catalys t and the
RI-4078C
47~9
amounts of hydrogen and normally gaseous hydrocarbons to be
handled by process equipment. During the cracking process, the
heavy metals transfer almost quantitively from the feedstock oil to
the catalyst particles and tend to deposit on interior and exterior
5 surfaces of the particles wherein they can block and/or retard
diffusion to active sites.
Since the various heavy metals are not of equal poisoning activity
relative to catalytic acid sites, it is convenient to express the
poisoning activity of an oil containing one or more of these metals
10 in terms of the amount of a single me-tal which is estimated to have
equivalent poisoning activity. Thus, the heavy metals content may
be expressed by the following formula in which the content of each
metal present is expressed in parts per million by weight based on
the weight of the oil: Nickel Equivalen-ts = Ni ~ V/4.8 + Fe/7.1 +
15 Cu/1. 23. In conventional FCC practice, crude oils are carefully
fractiona~ed to provide a gas oil with a relatively low level of heavy
metal contaminants, namely, 0 . 25 ppm Nickel Equivalents or less .
The above formula can also be used as a measure of the heavy
metals accumulated on the cracking catalyst itself, the quantity of
20 metal used in the formula being based on the weight of
moisture-free catalyst. In FCC practice, equilibrium catalyst is
removed and fresh, contaminant-free catalyst is added at a ra-te
sufficiently high to control the heavy metal conten-L of -the catalyst
at relatively low levels, namely, 1, 000 ppm Nickel Equivalents or
25 less.
Some crude oils contain from about 10% to about 30% by volume
of heavy hydrocarbons which will not boil below about 552C
~'0_5F) at atmospheric pressure. Atmospheric bottoms and vacuum
bottoms may contain even higher percentages of this highest boiling
30 fraction. ~he coke precursor and poison metal components of the
RI-4078C
4~7~9
crude are for the most part concentrated in this fraction.
Accordingly, many of the problems presented by these components
have been avoided by sacrificing the yield of liquid fuel fractions
which is potentially available from cracking the highest boiling
5 fraction. More particularly, in conventional FCC practice, the
crude oil has been vacuum fractionated to provide a FCC feedstock
boiling between about 343C (650F) and about 538C (1000F), this
fraction being referred to as a vacuum gas oil and being relatively
~ree of coke precursors and poison me tals . Vacuum gas oil is
10 generally prepared from crude by dis tilling off the fraction boiling
below about 343C (650F) at atmospheric pressure and then
separating by vacuum d;stillation one or more fractions boiling
be-tween about 343C (650F) and about S52C (1025F) from the
heaviest fraction boiling about 552C (1025F). The heaviest
15 fraction is normally not used as a source of catalytic conversion
feedstock, but instead is employed for o-ther purposes, such as the
production oE asphal-t, which represen-ts a was-te of the potential
value of this portion of the crude oil as a source of liquid fuels.
Due to the continually increasing demand for gasolines,
20 relative to heavier liquid fuels, coupled with shrinking supplies of
normally used gas oil cracking stocks, more attention has recently
been given to the catalytic cracking of heavier charges tocks, such
as residuals from which the highest boiling fraction has not been
separated. In addition, consideration has been given to blending
25 the heaviest or "resid" fraction with various lower boiling frac-tions
in order to increase overall conversion of crude oil to liquid fuels.
In view of the high potential value of the heaviest fraction of crude
oils, a number of methods have been proposed in the pas t to
overcome the problems associated with the cracking of feedstocks
30 contaminated with metals and coke precursors and thereby increase
the overall yield of gasoline and other hydrocarbon fuels from a
given quantity of crude oil. Sugges-t;ons have been made to
RI -407~C
447~
--6--
pretreat the contaminated feed to reduce the metals content to below
about 4 ppm nickel equivalents and the Conradson carbon residue to
below about 1. Various demetalization techniques have also been
suggested for removing the metal contaminants once they have been
deposited on the catalyst. Most of these prior art techniques,
however, require expensive additional equipment and materials can
cannot be justified from an economic standpoint.
Attention has also been given to developing improved catalysts
for cracking more contaminated feeds. However, many problems
have been encountered in the use of prior art catalys-ts for
cracking feeds containing resid fraction.
A catalyst comprised of crystalline zeolite particles embedded
within a larger matrix particle has numerous passages leading from
the outer peripheral surface of the matrix particle to the smaller
zeolite particles supported within the matrix. In this specification,
these matrix passages are referred to as "feeder pores". Feeder
pores in effect provide access passageways from the surface of each
catalyst particle to those zeolite particles at locations internal to the
matrix. There also may be a small but finite number of zeolite
particles exposed at the surface of the ma-trix.
Generally, the pores of zeoli-tic sieves fall within the range of
O O
4 to 13A. Accordingly, any pores larger than 13A are usually in
the matrix. In prior art catalysts of this -type, the average
diameter of feeder pores in fresh ca-talyst usually falls within the
range of about 30A to about 400A. Alumina-silica matrices generally
have pores in this r ange, although a relatively small proportion may
be larger. However, after an extended period of use, the effective
average pore diameter of these prior art catalysts may be decreased
significantly because of coke and metal accumulations. These prior
art catalysts have proven inefficien-t for cracking resid containing
RI -4078C
~24~
--7--
feedstocks for a number of reasons, including both a low zeolite
utilization factor and undesirable reaction diffusion limitations. The
low zeolite utilization effect is a consequence primarily of the
deposition of both coke and heavy metals in and/or across the
mouths of the working pores of the zeolite. These components in
effect block off zeolitic pore volume containing unused or
incompletely used acidic sites.
"Zeolitic pore volume" refers to the free volume of the
micro-pores in the zeolite component rather than the matrix. The
term "pore volume" as applied to the ca-talyst composition as a whole
refers to the free volume in the matrix and zeoli-te combined which
is provided by bo-th macropores (pores having a minimum diameter
above 30A ) and micropores (pores having a minimum diameter of
O O
30A or less). The pore volume fraction for pores greater than 30A
in diameter may be determined by mercury porosimetry methods,
such as the method of U.S. Patent No. 3,853,789.
The pore volume fraction in the 0 to 30A range may be
determined by the BET nitrogen adsorption method described by
Brunauer, Emmett, and Teller in the Journal of the American
Chemical Society, 60, 309 (1938). The pore volume of fresh
hydrocarbon conversion catalyst may vary widely depending upon
the size of the pores in the matrix and, where used, the relative
amount of catalytic promoter, such as zeolite, and the size of its
pores .
Diffusion limitations may result from a number of different
mechanisms. One such limitation is a consequence of high molecular
weight molecules in the feed and the absence of a sufficient number
of feeder pores of the size range required for transporting these
large molecules to the acidic sites of -the catalyst, some oE which
may be in the matrix but the majority of which are in the zeolite.
RI -4078C
ILZ~478~
--8--
Another diffusion limitation in the processing of reduced crude or
other resid con taining feeds is due to what may be called "pore
plugging". Pore plugging is caused by the absorption of
unvaporized hydrocarbons in the catalyst pores so that they are
5 impractical to remove by stripping operations prior to regeneration.
The trapping of heavy hydrocarbons which cannot be removed by
conventional stripping operations can lead to excessive coke and
regeneration temperatures and increased air consump-tion. Pore
plugging and the deposition of coke and/or heavy metals within or
10 over the pores also leads to decreased diffusion of reactants -to and
products from acidic si-tes. Slow diffusion rates may result in
thermal cracking predominating over cataly-tic cracking, which in
turn causes loss of selectivity. Thus, catalysts possessing
relatively small or restricted feeder pores will show relatively poor
15 cracking characteristics when cracking resid containing feeds,
including low conversion, poor selectivity, increased air
consumption during regeneration, and higher regeneration
temperatures. Hot spots also occur more readily during
regeneration and cause catalyst deactivation through sintering of
20 the matrix and loss of zeolite crystalline structure and acidity.
Furthermore, low catalyst utilization factors and diffusion limitations
both require high catalyst to oil ratios which necessitate relatively
low oil feed rates.
In order to provide economic levels of conversion activity and
25 more importantly the selectivity required in processing the very
refractory hydrocarbons found in resid fractions, it is desirable to
run the riser at a relatively high temperature. ln addition, large
amounts of coke accumulate on the catalyst. The primary problem
with this increased coke make is that the reactions in the
30 regenerator which conver-t coke to carbon monoxide and carbon
dioxide are highly exothermic. Since the regeneration reactions are
exothermic, the regeneration step is normally carried out at a
RI -4078C
~L2a~fl~7~19
g
temperature much above the cracking temperature in the riser.
This makes it necessary to run regenerator at maximum
temperatures in order to burn the coke off the catalys t to the
relatively low levels required for restoration of its activity. To
5 achieve a heat balance in cracking resid, it is therefore necessary
to operate the regenerator at very severe hydrothermal conditions,
which can cause rapid degradation of many prior art catalysts.
~ t high regenerator temperatures, excess heat and localized
hot spots may develop within catalyst particles, especially in places
10 where pore plugging has occurred or excessive coke deposits have
accumulated. These localized hot spots rasult in sintering and
collapse of the matrix pore s tructure, thus rendering a large
portion of the acidic sites in the matrix unavailable for further
reactant contact. Where a catalytic promoter is used, the promoter
15 will necessarily be entrapped within the collapsed pores of the
matrix and blocked off from further reactive contact. Coke from
resid molecules can also cover and block portal surface areas of
both the matrix and the zeolite.
The crystalline structure of zeolites is suscepti~le to
20 - degradation by high regenerator temperatures per so. Zeolites are
crystalline alumino-silicates made up of tetra-coordinated aluminum
atoms associated through oxygen atoms with silicon atoms in an
ordered crystalline structure . Localized ho t spots in or near the
zeolite par-ticles can cause destruction oE the aluminosilicate
25 crystalline structure, at least to the extent of destroying portal are
of the zeolite, with a resulting loss of its ca talytic action .
Furthermore, both sodium and vanadium contaminates accelerate
sintering and collapse of pore structures in both the matrix and
zeolite components. Such degradation permanently deactivates the
30 catalyst so that it must be removed from the system, resulting in
high make-up rates that may prove uneconomical because of the
RI -4Q78C
,
- ;~LZ~47~9
-10-
high coast of zeolite in the catalyst. There is a need therefore for
a heat resistant zeolite catalyst suitable for use in cracking resid
containing oil feeds with improved overall utilization of acidic sites
and a minimum of diffusion limitations.
While it has been recognized in the past that the physical
structure of ca talyst particles plays an important role in their
effectiveness, the ex-tent to which such structure is important
generally has been obscured by the lack of analytical -techniques for
isolating the complex mechanisms involved in catalytic cracking. In
-Lhis connection, some attention has been given in the prior art to
increasing the pore size of catalyst matrices. Thus, it has been
suggested that extremely large pores, such as those with diameters
above 1000A, might be introduced into a catalyst by incorporating
a removable material and subsequently removing that material during
catalys t preparation . See, for example, the catalyst described in
U.S. Patent No. 2,890,163 to Anderson, et al, and U.S. Patent No.
3,944,482 to Mitchell, et al . However, the removable materials
suggested from this purpose have not been easily controllable and
have resulted in poorly defined pore structures and a wide variety
of pore sizes, relatively few, if any, of the pore diameters being in
the actual size range needed for resid cracking.
Disclosure of the Invention
The present invention provides a catalyst and a process for
selective conversion of heavy hydrocarbons. It is therefore a
principal object of the invention to provide a special hydrocarbon
conversion catalyst resistant to deactivation by severe
hydro-thermal conditions and by accumulations of coke precursors
,,r,d poison metals. Another object of the invention is to provide an
improved process for cataly~ic conversion of high boiling
carbo-metallic oil feeds containing relatively high concentrations of
RI-4078C
~L2~47~9
coke precursors and poison metals. The catalyst and process of
the invention are particularly useful for cracking oil feeds which
contain significant quantities of residual hydrocarbons, e . g ., at
least ten percent, boiling above about 552C (1025F), and
significant quantities of heavy metals, e.g., at least about 4 ppm of
Nickel Equlvalents. Feeder pores having large minimum diameters
and large mou-ths are provided in the catalyst of the invention so as
to facilitate diffusion of high molecular weight molecules through the
matri~ to the portal surface area of the sieve particles. The
catalyst matrix also has a relatively large pore volume in order to
soak up unvaporized portions of the oil feed. The feeder pores
through the matrix are sufficiently large so that significant number
of hydrocarbon molecules can diffuse to active catalytic sites both
in the matrix and in the sieve. On the other hand, the open
channels of these feeder pores are somewhat tortuous and tend to
trap molecules having molecular weights of or greater and average
diameters of 200A or greater (for example asphaltenes, porphyrins
and polynuclear aromatics). These very large molecules can
effectively cover and block the relatively small pores of the zeolitic
sieve. The size and structure of the feeder pores of the invention
are such that they remain open without diffusion restrictions even
when matrix surfaces and pore channel walls become loaded with
very large asphaltene type molecules.
It has been discovered that the optimum average diameter for
feeder pores in the catalyst matrix is in the range of 400 to 6000A.
It has also been discovered that these pore size ranges can be
easily and readily obtained with the use of carbon black particles
having a cross-sectional diameter in the range of about 100 to 1000A
and a length to diameter ratio greater than about 2 :1, preferabl
greater than about 5 :1. Upon firing a catalyst composition
containing these carbon black particles, such as during the
regeneration cycle of a cracking process, the carbon burns out of
RI -4078C
~24~7~9
-12-
the matrix, leaving large feeder pores within the desired size
range. The carbon black may be added at almost any time during
the catalyst preparation prior to drying the final catalyst composite.
The fired catalyst has a relatively large and controlled number of
5 feeder pores within the desired size rar.ge, the number and size
distribution of these pores being enhanced to a significant ex-tent
by the type and amount of dispersant used for suspending the
carbon black in the aqueous suspension from which the unfired
composite is made.
The large feeder pores of -the invention extend from the
surface to the interior of the matrix particles and provide access
channels to the much smaller zeolite par-ticles supported within the
matrix material. Carbon black may be incorporated in the catalyst
matrix by adding it to a composite catalyst slurry along wi th the
15 zeolitic component prior to spray drying the slurry to form final
microspheres of matrix supported zeolite. Where a kaolin clay
composition is spray dried, calcined and then treated to generate
zeolite crystals in situ, the carhon black may be added to the clay
slurry fed to the spray drier so as to be present in the composition
20 during and after in situ formation of the zeolite within the matrix
material .
A competing consideration for the selectivity desired in
cracking residual feeds is that there must be sufficient acid sites
present so that catalytic cracking dominates the conversion
25 reaction. Thus, as average pore size and pore volume increase,
catalytic surface area may decrease. To compensate for this
decrease in surface area, the amount of zeolitic promoter can be
increased and/or smaller zeolite particles ~crystals~ employed to
increase the portal surface area of this component. A preferred
30 catalyst of the invention therefore contains relatively high
concentrations of very small æeolitic sieve par-ticles supported within
.
RI -4078C
~za~47~3~
-13-
an alumina, silica and/or alumina-silica matrix. Both the sieve and
the matrix should have good steam and thermal stability. A "Y"
type zeolite sieve relatively free of sodium and stabilized with
hydrogen and/or ammonium ions and/or rare earth ions is
5 preferred. According to a preferred embodiment of the invention,
spray dried microspheres containing carbon black are partially
exchanged with rare earths, calcined to remove the carbon black
and stabilize the zeolite, and further exchanged with rare earths to
provide a catalyst having superior hydrothermal stability.
Acidic sites may also be provided in the matrix material so that
at leas t some of the heavier hydrocarbons, both liquid and
vaporous, can be cracked on the surfaces and in the passages of
the matrix to provide a means by which molecules larger than the
sieve pores can be converted to smaller molecules of a size small
enough to enter and be cracked in the sieve. For example, alkyl
fragments can be stripped from large aromatic molecules so that the
fragments may enter the highly active pores of the zeolite.
The large feeder pores and process conditions of the present
invention provide increased conversion and improved selectivity in
the cracking of reduced crude and other resid containing oil feeds.
In this connection, it is believed that these larger pores are
capable of maintaining adequate diffusion of reactants and reaction
products while providing sufficient pore volume and surface area in
the matrix for absorption and retention of poison metals, coke and
unvaporizable hydrocarbons. Although not wishing -to be bound by
any particular theory or hypothesis as to the reasons for the
improvements afforded by the invention, the following matters may
contribute to its success.
According to the literature, a normal C45 hydrocarbon molecule
has a boiling point above 815C (1500F~, a cross-sectional diameter
RI -407~C
~2~-~7~9
-14-
O O
of about 4A and a maximum length of about 50A. Therefore, with a
conventional catalyst matrix having an effective pore diameter at
equilibrium of less than 50A, this heavy molecule can enter the
average pore based on its minimum cross-sectional dimension but
5 cannot enter based on its length dimension. Thus some net
orientation would be required for heavy molecules of a resid
fraction to traverse the pores of the matrix and reach the acid sites
of the zeolite. Statistically, heavy molecules, such as those having
molecular weights in the range of 1000 to 10, 000, would exist in a
10 relatively disordered state so that some of the molecules would not
be able to enter the matrix of catalysts having relatively small
feeder pores. Since the reactants and the products of catalytic
conversion must necessarily use the same feeder pore passages and
diffuse in opposite directions, the rate of diffusion into a pore must
15 equal the rate of diffusion out under steady state conditions. It is
therefore believed that the large feeder pores of the invention
decrease the necessity for a particular molecular orientation and
therefore increase diffusion rates through these pores, particularly
at equilibrium conditions where effective pore diameters necessarily
20 reflect deposits of coke and heavy metals on the wall of pore
channels .
The larger feeder pores of -the matrix allow coke and metal
deposition near the surface of the catalyst particles without pore
blockage and absorption of heavy liquid hydrocarbons without pore
25 plugging. Large vaporized hydrocarbon molecules can enter and
exit these pores at equilibrium conditions without special orientation
and can therefore more easily reach acidic sites which are
concentrated in the zeolite but also may be present to a lesser
extent in the matrix material. In a sense, the large pores are able
3 to "soak up" both metal poisons and liquid coke precursors and in
effect neutralize at least a portion of these contaminants. The high
pore volume provides physical space in which to "load" there types
RI -4078C
~LZ~L47~39
-15-
of contaminants without unduly restricting diffusion, even in the
presence of partial sintering. The large pores also facilitate
reactions between the trapped material and elements or compounds
that may be added to the riser or regenerator in order to passivate
5 the poison metals accumulated on the matrix.
Unusually high temperatures are needed to crack resid
containing feedstock for the reasons given previously. However, at
the high temperatures proposed, thermal cracking reactions compete
with catalytic cracking reactions. The product distribution
10 (selectivity) for -the thermal reaction is quite different from the
catalytic reaction and highly undesirable, the thermal reaction
yielding much lighter gases, more coke, high boiling gas oils with
high Conradson carbon values, and relatively low octane gasoline
range products. The activation energy for catalytic cracking is
15 considerably lower than for thermal cracking. With prior art
catalyst, as the coke and metals deposit on the catalyst and the
liquid asphaltene components fill the relatively small matrix pores,
diffusion of the hydrocarbon reactants to the acidic sites of the
zeolite is retarded to the extent that diffusion through the matrix
20 becomes rate limiting. As this occurs, the apparent activation
energy for catalytic cracking rises and diffusion becomes rate
limiting, as this occurs, the activation energy for thermal cracking,
which remains cons-tant, results in greater proportions of the
undesirable thermal produce distribution. Because the hydrocarbon
25 molecules of a residual feed have relatively easy access to the
zeolite in the catalyst disclosed, the catalytic reaction is much
faster than the thermal reaction, giving higher conversion and the
desired catalytic product distribution.
In addition, improved diffusion rates for regeneration gases
30 and combustion products within the particles help avoid hot spots
during catalyst regeneration and thereby reduce sintering of both
Rl-407~C
~L2~
-16-
the matrix and the promoter materials. Coke also tends to deposit
hearer to the particle surface where it can be more easily reached
by regeneration gases and burned off at lower temperatures.
While the pore volume is large, the surface area of the matrix
5 may be correspondingly low. This relatively low surface area,
together with large feeder pores, minimizes the physical and/or
chemical retention of vaporizable hydrocarbon molecules so as to
facilitate stripping these molecules from the catalyst and reducing
the amount of coke carried into the regenerator. The amount of
10 carbon burning is less and the amount of air necessary for
regeneration is -thereby reduced.
Although the carbon black (CB) catalyst disclosed may be used
in a variety of conversion processes employing a wide variety of
contacting equipment, it is particularly useful in the catalytic
15 cracking apparatus an process of the invention. The apparatus
disclosed comprises a progressive flow riser with a ballistic
separator at the upper end of the riser for causing a sudden and
substantially instantaneous separation of ca-talyst particles from
product vapors. The catalyst is then transferred to 2 stripping
20 vessel for removal of residual hydrocarbons removable in the
presence of high temperature steam and/or other stripping gases.
Stripped catalyst is then -transferred to a two-stage combustor
having insufficient oxygen in the first stage to convert all of the
carbon to carbon dioxide and an excess of oxygen in the second
25 stage to almost completely burn off the carbon remaining after the
first stage. The regenerated catalyst particles attain relatively
high temperatures in the range of about 704C (1300F) to 815C
(1500F) and have very low levels of residual carbon, namely, 0.05
weiaht percent or less. The regenerated catalyst particles are then
30 returned to the bottom of the riser where they are contacted wi th
fresh feedstock.
RI-4078C
~24~789
-17-
The conversion process of the invention is particularly
effective in utilizing the CB catalysts for cracking resid containing,
carbo-metallic feeds of the type described. Fresh oil feed and a
diluent, such as steam, are mixed with -the hot regenerated
5 cracking catalyst at or near the bottom of the riser, vapori7ing
and/or fluidizing the feed and diluent substantially instantaneously
to form a vaporous suspension that flows rapidly upward to the
ballis-tic separator . The temperature and ca talys-t oil ratio in the
suspension is sufficiently severe to convert approximately 50 to 90%
10 of the carbo-metallic oil feed to gasoline per pass and produce
relatively high levels of coke on -the catalyst. The velocity of the
suspension combined with ballistic separation is such as -to provide
very short contac t times and avoid overcracking the desired
molecular species of the product, notwithstanding the high
15 temperatures and very active catalyst. The regenerator also is
operated at relatively high temperatures which provides rapid and
effective coke removal and the heat necessary for the endothermic
riser reaction. In view of the high temperatures, the regenerator
configuration is such that -the average catalyst hold up time for
20 regeneration is relatively short, namely, on the order of about 3 to
5 minutes or less. Since relatively small amounts of catalyst are
held up in other portions of the system, the overall catalyst
inventory is significantly low.
Relatively high oxygen partial pressures are maintained in the
25 regenerator, either in the last stage of multistage regenerators or
in a zone immediately upstream of discharge conduits from single
stage regenerators to keep heavy metals on the catalyst in their
less active oxide form. The substantially instantaneous Eluidization
of the oil feed and the very short residence times employed tend to
30 inhibit reduction of these metals in the riser to their more
catalytically active free metal state . I t is also con-templated that
certain elements, such as antimony, may be added to the
RI -4078C
4789
-18-
regenerator or riser to more permanently tie up accumulated heavy
metals. The large pores and smaller surface areas of the CB
catalysts facilitate such metals deactivation reactions. Although
economically prohibitive at the present time, demetalization
5 techniques also should prove to be more effective with CB
catalys-ts .
Because the catalyst is resistant to degrada-tion and process
apparatus and parameters are adapted for its effective utilization,
the make-up rates at which fresh catalyst must be in-troduced into
10 the system are well within accep-table limits, namely, in the range of
about 0.1 to 3.0 pounds of catalyst per barrel of fresh feed. ~lore
particularly process parameters of the invention, together with
further par-ticulars on the catalyst and apparatus thereof, are given
in the description below of the best mode for carrying out the
15 invention.
Brief Description of the Drawings
The invention may be further understood by reference to the
description of the best mode taken in conjunction with the
accompanying drawings in which:
Figure l is a diagrammatic illustration of the reticulated
structure of one type of carbon black employed in the catalyst
composition of the invention.
Figure 2 is a diagrammatic representation of a catalyst particle
made according to he invention.
" Figure 3 is a diagrammatic representation of a feeder pore of
the invention enlarged relative to -the view of Figure 2.
RI-'1078C
~L2gL47~9
-19-
Fi0ure 4 is a diagrammatic representation of cracking a
polynuclear aromatic hydrocarbon within the matrix of the catalyst
of the invention.
Figure 5 is a diagrammatic representation of cracking the
5 aliphatic fragment of Figure 4 in -the zeolitic component of the
catalyst of the invention.
E'igure 6 is a graph showing changes in pore volume ranges
rela-tive to changes in the amount of carbon black used in making
the catalyst of the invention.
Figure 7 is a schematic diagram of an apparatus for utilizing
the catalyst and carrying out the process of the invention.
Figure 8 is a schematic diagram of another apparatus for
utilizing the catalyst and carrying out the process of the invention.
Figures 9A, 10A, 11A, and 12A are micrographs made by a
15 scanning electron microscope (SEM) showing reticulated carbon
black particles of the type employed in the catalytic composition of
the invention.
Figures 9B, 10B, 11B and 12B are contour plots and other
data from photo-grammetric topography and correspond to Figure 9A
20 through 12A, respectively.
Figure 13 is a SEM micrograph showing a catalyst matrix made
according to the invention.
Figure 14 is a graph showing the change in catalyst relative
activity with decreasing volume % MAT conversion.
RI -4078C
~z~ 9
-20-
Best Mode for Carrying Out the Invention
The Feedstock
The catalyst of the invention may be used for treating any
hydrocarbon feedstock suitable for cracking to lower boiling
5 components or fore reforming or other hydrocarbon conversion
processes. It is especially useful for cracking oil feeds containing
an appreciable amount of high molecular weight components, for
example a feedstock with at least S weight percent, preferably at
least 10 weight percent, not boiling below 552C (1025F).
10 Diffusion limitation problems, in general, become increasingly
troublesome with increasing fractions of high molecular weight
components in the feedstock. Other factors involved include
molecular configuration and the like. The terms "high molecular
weight" and/or "heavy" components refer to those hydrocarbon
fractions having a normal boiling point of at least 552C (1025F)
and include non-boiling hydrocarbons, i.e., materials which may not
boil under any conditions.
These heavy components have relatively large molecular
dimensions, the largest average dimension of each component ~eing
referred to here as its "average diameter". For purposes of
selecting a catalyst having an optimum most frec~uent feeder pore
size and distribution, some knowledge of -the range of average
diameters of the feedstock is desirable. One method of evaluating
molecular sizes of large organic molecules is that of J. J. Hermans,
et al (J . Chem ~ Phys ., Vol . 20 , page 1360 , 1952~ , in which the
average effe~tive molecular diameter in Angstroms (A) is equal to
7 . 4 N1/2 where N is the number of carbon atoms in the molecule .
Since various molecules have a variety of configurations and
shapes, this procedure gives a relative indication of molecular sizes
and provides a useful tool for estimating average molecular sizes
RI-4078C
~LZ~47~
-21 -
and spreads oi sizes in feedstocks. Such data may be used for
selecting the predominant feeder pore sizes and size distribution
characteristics of the catalyst to be prepared for given types of
feedstocks. This method therefore may be used as the basis for
calculating average molecular diameters and ranges of diameters for
heavy feeds. Generally, high boiling hydrocarbon feedstock has a
relatively wide range of molecular diameters for the molecules
present with significant frequency. For example, the average
difference between the lower and higher significan-t diameters may
be as high as 250A . I t is preferred that the ratio of average
feeder pore size of the catalyst to average molecular diameter of the
552C (1025F) portion of the feedstock be at least about 2,
preferably at least about 5.
The composi-te catalysts disclosed have a high tolerance both to
metals and to coke precursors and these catalysts will economically
crack feeds tocks containing high concentrations of such
contaminants. A high metals feedstock for purposes of this
invention is one having a heavy metal content of at least about 4
ppm of Nickel Equivalents. A high coking feedstock for purposes
of this invention is one having a Conradson carbon residue value
greater than about 1. The feedstocks for which the invention is
particularly useful will have a heavy metal content of at least about
5 ppm Nickel Equivalents and a Conradson carbon residue of at
least about 2. The greater the heavy metal content and the
Conradson carbon residue, the more advantageous the ca talys t and
process of this invention becomes. A particularly preferred
feedstock for treatment by the process of the inven-tion includes a
reduced crude and comprises 70 percent or more of a 343F+
(650F+) material having a fraction greater than 20 percent boiling
above 552C (1025F) at atmospheric pressure, a metals content of
greater than 5 . 5 ppm Nickel Equivalents and a Conradson carbo
residue of greater than ~. This feed may have a hydrogen to
RI-~1078C
~ 2~4789
-22-
carbon ratio of less than about 1. 8 and coke precursors in an
amount sufficient to yield about 4 to 14 percent coke by weight
based on fresh feed. If the metals and/or Conradson carbon are
higher than these values, the feed may be pre~reated (but
5 preferably is not) by a hydrotreating step to saturate unsaturated
hydrocarbons and/or by contacting adsorbent particles to remove a
portion of the poison metals and carbon precursors.
Representative feedstocks contemplated for use with the
invention include whole crude oils, fractions of crude oils such as
10 topped crude, reduced crude, vacuum fractionator bottoms and
other fractions containing heavy residua, coal-derived oils, shale
oils, waxes, untreated or deasphalted residua, and blends of such
fractions with gas oils and the like.
In addition to feedstock per se, added diluent materials may
15 also be charged to the riser to lower the vapor pressure of the oil
feed. Diluents will increase the space velocity of the process by
accelerating the velocity oE the oil and decreasing catalyst contact
time. Any diluent which is a vapor or becomes a vapor under the
conditions in the conversion zone can be used. If the diluent is a
20 hydrocarbon, is should desirably have a boiling point below about
343C (650F), and more preferably it should be a gasoIine range
h-ydrocarbon, e . g. naphtha, or lighter, which fractions boil at
about 221C (430F) or below. If added hydrocarbon boils above
343C+ (650F+), it will itself be considered a portion of the
25 cracking feedstock. Other diluents include various gases such as
hydrogen, nitrogen, methane and ethane, and water, which may be
charged either as liquid or steam. Such diluents may be added a-t
or near the bottom of or at one or more locations along the riser
~Gnversion zone so as to assist in dispersal and fluidization of the
30 catalyst, dispersal and vaporization of the liquid feedstock,
quenching of the catalys-t and~or oil suspension, and und~r some
RI-4078C
~ .
:~2~7~9
-23 -
conditions, may increase the cracking rate and/or improve the
selectivity of the cracking process.
With respect to the tolerance levels of the catalyst itself, the
heavy metals may accumulate on the ca talyst to levels in the range
of from about 3000 to about 20,000 ppm of Nickel Equivalents .
Where about 4 -to 14 percent of the feeds-tock is converted to coke,
-this coke is generally deposited on the catalyst in amounts in the
range of about 0.3 to 3.0 percent by weight of -the catalysts.
The Catalyst
The present invention includes a method of making a catalyst
with a relatively large percentage of pores in size ranges above
o O
100A, preferably in the range of 400 to 6000A, more preferably 1000
to 6000A. The catalyst is especially useful in the processing of
reduced crudes and o-ther heavy feeds. The method involves
incorporating a relatively large amount of selected carbon black
solids into the catalyst matrix material. The carbon black is
subsequently removed by converting it to gaseous carbon oxides by
oxidation at elevated temperatures. Its removal provides a pore
volume greater than 0.10 cc/gm in pores greater than 400A,
preferably 0.15 cc/gm with at least 0.10 cc/gm in pores greater
than 1000A, more preferably at least 0.20 cc/gm with at least 0.15
cc/gm in pores greater than 1000A. Relative to the prior art, the
invention reliably and predictably increases the pore volume
comprises of these large pore sizes.
Carbon black may be added, along with other solids such as
zeolite and clay, during the formation of the ca talyst matrix
material. Alternately, the carbon black may be added after the
basic catalytic composition has been prepared but before final
drying, as for example, by forming a slurry of prepared catalyst
RI-407~C
SL2~7~9
-24-
composition followed by dispersing the carbon black within the
slurry and then spray drying the resulting suspension. Carbon
black may also be added to the zeolite component either prior -to or
during the formation of the aluminosilicate crystals.
Carbon black differs from other types of carbon, such as
charcoal, and from other types of fillers such as flour or cellulose
fibers, in that it exists as very small reticulated particles and is
substantially non-porous in mos-t of its forms. Each retlculated
particle is itself comprised of smaller "primary" carbon black
particles. Primary carbon black particles are essentially spherical,
the diameter of the sphere varying depending upon the method of
manufacture. The primary particles in turn are composed of
several thousand microcrystallite bundles stacked together in a
random order and each bundle consists of several polynuclear
aromatic platelets which are stacked in a not quite parallel manner.
The carbon blacks of the present invention possess to a greater or
lesser extent a basic property called "structure". Struc-ture refers
to the degree to which the primary particles are bound together
into a 3-dimensional primary chain network making up the
reticulated particle. An idealized reticulated particle of carbon
black is illustrated in Figure 1. While not intending to be bound
by any one theory, i-t is believed that the primary particles may be
fused together or share common microcrystallite bundles or planes
to make up the primary chains. This primary reticulated s-tructure
is to be distinguished from secondary or reversible structures
which result from van der Waals forces between individual
reticulated particles. Primary carbon black structures exhibit a
pronounced tendency to agglomerate into secondarv reticula ted
structures when dispersed in almost all media.
The carbon black of the present invention is preferably
produced in refractory-lined furnace reactors by pyrolysis of highly
RI-4078C
~Z~147B9
-25-
aromatic refinery by-product oils. These oils are subjected to
temperature of about 760C (1400F) to 899C (1650F) in a reaction
zone maintained at conditions producing an endothermic reaction
which strips atomic hydrogen from the aromatic hydrocarbon
5 molecules to leave aromatic carbon nuclei. The resulting reticulated
particles in the form of a black l'smoke" are quenched in a
downstream tunnel by water injection at a point several feet from
the reaction zone. In this method of manufacturing carbon black,
the primary par-ticles size can be closely controlled and produces
particle diameters in the range of about 200 to 900A. These
primary particles are simultaneously bound together to form primary
reticulated chains having lengths in the range of about 500 to
30, OOOA . Both structure and particles size may be closely
controlled through the design of the oil injection nozzle, reaction
15 chamber geometry, pyrolysis temperature, residence time, and the
intensity of gaseous turbulence.
Another important feature of carbon black is that it contains
less than about half the amount of hydrogen theoretically needed to
bond all edge portions of the polynuclear aromatic platelets and it is
20 believed that the particles contain many unsatisfied valences or free
radicals. A number of elements, such as oxygen and sulfur, may
interact at some of the peripheral positions of -these micrographic
platelets so as to form complexes which are generally analogous to
the functional groups of organic compounds. The principal surface
25 groups on carbon black have been have been identified in the
literature as carboxylic acid, phenolic hydroxyl and quinone
groups, and possibly peroxide and lactone groups. In addition to
accepting or trapping free radicals, it has been suggested that
functional groups at the carbon black surface can also generate free
30 radicals, or at least can initiate free radical reactions. It is
believed that these organic functional groups may assist in Eorming
both the unfired catalyst compositions of the present invention.
RI-4078C
~Z447~39
-26-
It should also be noted that carbon black is an amorphous form
of carbon as opposed to graphite which is a soft crystalline form of
carbon that differs greatly in properties from amorphous carbon.
The thermal conductivity of amorphous carbon is relatively high and
5 is equivalent to some metals. Carbon black also has a very low
co-efficient of thermal expansion and a high resistance -to thermal
shock. It is believed that these features contribute to relatively
rapid rates of carbon burnout and large feeder pores of
substantially uniform diameter.
10A preferred t~pe of carbon black meets the specifications of
ASTM No~ N-219. These blacks have relatively low structure and
are made using an intermediate super-abrasion furnace. Such
blacks are available from Ashland Chemical (United N-210), Cabot
(Regal 600), Columbian (Niotex 130), Continental (Continex
15ISAF-LS), and Phillips (Philblack N-210). United N-219L is
preferred as it is not compacted but supplied loose at relatively low
bulk density compared to pelleted or compacted blacks.
The average primary particle diameter of this black is about
300A and it has an ASTM Iodine Number of about 115 (a
20 measurement of surface area per unit weight correlating well with
nitrogell absorption measurements for furnace blacks). The
relatively low structure of this black is indicated by a low DBP
Absorption value of about 0.78 cubic centimeters per gram (the DBP
Absorption value is indicative of the degree of linkage between
25 primary carbon black particles).
One of the principal objec-ts of the invention is to disperse the
carbon black sufficiently in the catalyst forming media so that
formation of secondary CB structures through agglomeration of
primary CB structures can be significantly controlled. Some of the
30 more effective dispersants for carbon black in in aqueous media are
RI -4078C
. . .
~L2~478~
--27-
hexadecyltrime-thylammonium bromide, an ethoxylated alcohol sulfate
sold under the brand name Marasperse CBO-3, and mixtures
thereof. These dispersants are used in amounts generally
proportional to the weight of carbon black added, preferred
proportions being in the range of about 0.05 to 1.0 weight percent
of carbon black, more preferably about 0.1 percent. Quaternary
surfactants such as ~uaternary O, succinates such as Aerosol, and
other ethoxylated alcohol sulfates may also be used.
With carbon black, it has been found that the most
predominant feeder pore size can be controlled to a significant
extent both be the amount of carbon black used and the amount and
effectiveness of -the dispersant used to suspend the carbon black in
an aqueous medium. It is believed that the degree of dispersion
versus the degree of aggl~meration of the primary reticulated CB
particles is a controlling factor in determining whether the
predominant feeder pore sizes are in the lower or the upper portion
of the preferred pore size range of 500 to 6000A. Thus, lower
carbon black concentrations in combination with the most effective
dispQrsants provide an increased number of pores with effective
diameters in the range of 400 to 1000A when using primary carbon
black particles having an average diameter of about 300A. Larger
concen-trations of carbon black and lower and/or less eEfective
concentrations of carbon black dispersants provide predominan-t
feeder pore sizes in the range about 1000A. This phenomenon is
believed to be due to the formation of feeder pores by agglomerates
of two or more individual reticulated CB particles.
The amount of carbon black used in preparation of the unfired
catalyst will therefore depend on the extent to which large feeder
pores are desired in the final catalyst structure. Other factors
include the final attrition resistance desired. Generally, the amount
of carbon black should be in the range of 1 to 35 percent by
RI -4078C
.
~L2~7~
-28-
weight of unfired product. Too little carbon black will not produce
a sufficient number of large pores and too much carbon black will
result in a catalyst having relatively low attrition resistance.
Accordingly, preferred amounts of carbon black are in the range of
about 2 to 30 percent by weight, more preferably 5 to 15 percent
by weight, of the unfired product.
Another fac-tor to be considered in selecting the amount and
type of carbon black is the average effective diameter of the
hydrocarbon molecules in the feedstock. The average diameter
1~ referred to here is the statistical average of the lar~est e~fective
dimension of the molecules boiling above 552C (1025F). The
feeder pores should have an average effective diameter at least
equal to this average feedstock dimension, but should not exceed
about 10 times this dimension so as not to decrease unduly the
surface area and, correspondingly, the number of catalytic sites
available for the cracking reaction. Accordingly, the ratio of
average feeder pore diameter to average hydrocarbon diameter
should be in the range of 2 to 10 more preferably 4 to 8, most
preferably 5 to 8.
Another important feature of the invention is that the carbon
black suspension, preferably an aqueous medium, can be mixed
uniformly with a catalyst slurry and the resulting composite
suspension spray dried to form substantially uniform microspheres
within the preferred range of particle sizes described below. Be
comparison, prior ar-t techniques using a decomposable solid for
introducing large pores into a matrix involve the formation and
extrusion of a viscous paste which then has to be dried and broken
up. This results in catalyst particles having a wide range of
di~.rerse shapes and sizes which have to be sifted in order to
provide a catalyst of any uniformity. In addition, the types of
decomposable solids used in the prior art produce low ac tivity
RI-4078C
LZ~478~
-29-
catalysts with excessively ]arge pores, the pore size range in any
given mix being virtually uncontrol]able.
After being intimately mixed with the matrix material and any
other ingredients, such as zeolite and/or filler components, the
5 composite is shaped and dried to produce an unfired catalyst
composite. This shaped composite is then heated to burn out the
carbon black and produce a final catalyst product containing a
significant volume of large feeder pores within the desired size
ranges. The temperature experienced by the catalyst particles
10 should not cause objectionable changes in the structure of either
the zeolite or the supporting matrix. Where carbon hlack is
removed during manufacture, burn out is initiated at about 260C
(500F) and the firing time varies in accordance with the
temperature selected, higher temperatures requiring shorter firing
15 times. Where carbon black is burned out in the process unit
during regeneration, the temperature should not exceed about 815C
(1500F) to avoid damage to the zeolite. Preferred firing
temperatures are in the range of about 538C (1000F) to 787C
(1450F), with corresponding firing times from about three hours to
20 as low as a few minutes, such as associated with catalyst hold-up in
a regenerator. Where carbon b]ack is burned out of the
composition prior to in situ formation of the zeolite, temperatures as
high as 1093C (2000~F) may be tolerated.
The invention is not restricted to the use of contact agents
25 containing any specific matrix components or catalytic promoter.
Any matrices and/or promoters of the prior art may be used in
combination with carbon black for the production of feeder pores in
solid catalysts which may be of any suitable shape and size. For
example, carbon black may be used to provide feeder pores in
30 synthetic silica-alumina catalysts of the type described in U. S.
Patent ~o. 3,034,994 to Braithwaite, et al,
RI -4078C
. , , ~
7~
-30-
However, it is preferred to use
this or a simi]ar silica-alumina composition as a matrix for
supporting a superactive zeolite component. It is to be further
understood that other promoters, such as catalytically active metals
5 or metal compounds, may be used in place of or along with a
~eolitic promoter. A preferred catalyst of the present invention
therefore comprises three main components, namely, a catalytically
active or inactive matrix material, a superactive catalytic promoter
dispersed in the matrix material, and a carbon b]ack initially
10 dispersed in the matrix but removable therefrom by combustion.
Virtually any refractory oxide material capable of maintaining
stable pore characteristics may be used as the matrix for the
catalyst of this invention. A preferred matrix composition is one
having sufficient acid sties to provide significant cracking activity,
15 particularly for the high molecular weight components of the ~eed.
It is therefore a further object of the invention to employ a catalyst
matrix in which significant catalysis of the heavy hydrocarbon
molecules boiling above 552C (1025F) is affected in the feeder
pores. Catalysis in these macropores may be affected by acidic
20 sites either in the matrix itself or on exposed outer surfaces of the
superactive zeolite. It is believed that conversion and selectivity is
significantly improved if these feed components are init,ally cracked
in the matrix to smaller size molecules capable of entering the much
smal]er zeolite pores. Less reliance on thermal cracking for these
25 types of reactions gives improved product distribution and yasoline
yield and better overall product quality, i.e. Iess methane, ethane,
ethylene, thermal coke, thermal gas oil, and thermal gasoline.
A catalyst microsphere made according to the invention is
illustrated in Figure 2. The mechanism for trapping heavy liquid
30 asphaltenes in the matrix feeder pores so as to reduce the number
reaching and blocking zeolitic pores is illustrated in Figure 3.
Rl-4078C
..
~Z~4789
--31--
Figure 4 shows diagrammatically the cracking of a heavy polynuclear
aromatic at an acidic site (H+) in the matrix. The straight chain
fragment from the matrix cracking of Figure 4 may then be cracked
and reformed by the zeolite as illustrated in Figure 5.
The matrix material should possess good hyrothermal stability.
Examples of materials exhibi-ting relatively stable pore
characteristics are alumina, silica-alumina, silica, silica-magnesia,
magnesia-alumina, silica-zirconia, clays such as kaolin, metakaolin,
halloysite, anauxite, dickite and/or macrite, and combinations of
these materials. Other clays, such as natural montmorillonite,
synthetic mica montmorillonite (S~), and/or pillared layered clays
(PLC) may be added -to increase the acidity of the matrix. Clay
may be use in its natural or thermally modified states. The
preferred matrix of U . S . Patent No . 3,034,994 is a semisynthetic
combination of clay and silica-alumina. Preferably the clay is
mostly a kaolinite and is combined with a synthetic silica-alumina
hydrogel or hydrosol. This synthetic component forms preferably
about 15 to 75 percent, more preferably about 20 to 25 percent, of
the fired catalyst in weight. The proportion of clay is such that
the catalyst preferably contains after firing about 10 to 75 percent,
more preferably about 30 to 50 percent, clay by weight. A most
preferred composition of the matrix contains approximately twice as
much clay as synthetically derived silica, alumina or silica-alumina.
Synthetically derived silica-alumina should contain 55 to 95 percent
by weight of silica ~SiO2), more preferably about 75 percent.
After introduction o the zeolite and/or other promoters, the
composition is preferably slurried and spray dried to form catalys t
microspheres. The particles size of the spray dried matrix is
generally in the range of about 5 to 160 microns, preferably 50 to
80 microns.
RI -4078C
- :~LZa~47~3~
-32-
Various processes may be used in preparing the synthetic
silica-alumina, such as those described in U. S. Patent No.
3, 034, 994 . One of these processes involves gelling an alkali metal
silicate with an inorganic acid while maintaining the pH on the
5 alkaline side. An aqueous solution of an acidic aluminum salt is
then intimately mixed with the silica hydrogel so tha-t the aluminum
salt solution fills the silica hydrogel pores. The aluminum is
thereafter precipitated as a hydrous alumina by the addition of an
alkaline compound.
As a specific example of this method of preparation, a silica
hydrogel is prepared by adding sulfuric acid with vigorous agitation
and con-trolled temperature and concentration conditions to a sodium
silicate solution. Aluminum sulfate in water is then added to the
silica hydrogen with vigorous agitation to fill the gel pores with -the
15 aluminum salt solution. An ammonium solution is then added to the
gel with vigorous agitation to precipitate the aluminum as hydrous
alumina which combines with the silica to produce silica-alumina on
the surface of the silica hydrogel, after which the hydrous gel is
processed, for instance, by separating a part of the water on
20 vacuum filters and then drying, or more preferably, by spray
drying the hydrous gel to produce microspheres. The dried
product is then washed to remove sodium and sulfate ions, either
with water or a very weak acid solution. The resulting product is
then dried to a low moisture content, usually less than 25 percent
25 by weight, e.g., 10 percent to 20 percent by weight, to provide
the finished catalyst product.
The silica-alumina hydrogel slurry may be filtered and washed
in gel form to affect purification of the gel by the removal of
dissolved salts. This may enhance the formation of a continuous
30 phase in the spray dried microspheric particles. If the slurry is
prefil tered and washed and it is desired to spray dry the filter
RI-4078C
_33_ ~24~a789
cake, the ]atter may be res]urried with enough water to produce a
pumpable mixture for spray drying. The spray dried product may
then be washed against and given a final drying in the manner
previously described. Spray dryable compositions to which carbon
5 black can be added and spray drying techniques usable with the
present invention are also described in U.S. Patent No. 3,867,308
to Elliott and U.S. Patent No. 4,126,579 to F]aherty, et al,
Suitable catalytically active promoters other than zeolites may
10 be used and include metals or a catalytic compounds of metals such
as Pb, Sn, Bi, Ge, Sc, Ti, Cr, Mn, Co, ;Zn, Y, Nb, MQ, Ma, Ru,
Rh, Pd, La, ~-lf, Ta, W, Re, Os, Ir, Pt, Zr, Ac, Th, Pa, and U
and the like. These components may be used alone or in addition
to a superactive zeolite. In the latter case, these elements and/or
15 compounds may increase the catalytically active sites available in the
matrix. These additional promoters may be used in concentration
ranges generally from about 0.5 percent to about 20 percent, more
preferably about 1 to 5 percent by weight of fired catalyst.
The catalytically active promoter for à preferred catalyst
20 composition is a crystalline aluminosilicate zeolite, cormnonly known
as mo]ecular sieves. Molecular sieves are initially formed as alkali
metal aluminosilicates, which are dehydrated forms of crystalline
hydrous siliceous zeolites. However, since the alkali form does not
have appreciable activity and alkali metal ions are deleterious to
25 cracking processes, the aluminosilicates are ion exchanged to
replace sodium with some other ion such as, for example, ammonium
and/or rare earth metal ions. The silica and alumina making up the
structure of the zeolite are arranged in a definite crystalline
~attern containing a large number of small uniform cavities
30 interconnected by smaller uniform channels or pores. The effective
size of these pores is usually between about 4A and 13A.
RI -9078C
~ "
-34~ ~LZ~47~3~
The zeolites which can be employed in accordance with this
invention include both natura] and synthetic zeolites. The natural
occurring zeolites include gmelinite, clinopti]olite, chabazite,
dechiardite, faujasite, heulandite, erionite, analcite, levynite,
5 sodalite, cancrinite, nehpeline, Icyurite, scolicite, natrolite,
offretite, mesolite, mordenite, brewsterite, ierrierite, and the like.
Suitable synthetic zeolites include zeolites Y, A, L, ZK-4B, B, E,
F, H, J, M, Q, T, W, X, Z, ZSM-types, a~pha, beta and omega.
The term "zeolites" as used herein contemplates not only
10 a~uminosilicates but substances in which the aluminum is replaced by
gallium and substances in which the silicon is rep~aced by
germanium .
The zeolite materials utilized in the preferred embodiments of
this invention are synthetic faujasites which posses silica to alumina
ratios in the range from about 2.5 to 7.0, preferably 3.0 to 6.0 and
most preferably 4 . 5 to 6. 0. Synthetic faujasites are widely known
crystalline aluminosilicate zeolites and common examples of synthetic
faujasites are the X and Y types commercially available from the
Davison Divison of W. R. Grace and Company and ~he Linde
Division o~ Union Carbide Corporation. The ultrastable hydrogen
exchanged zeolites, such as Z-14XS and Z-14US from Davison, are
particularly suitable. In addition to faujasites, other preferred
types of zeolitic materials are mordenite and erionite.
The preierred synthetic faujasite is zeolite Y which may be
prepared as described in U . S . Patent No . 3 ,130, 007 and U . S .
Patent No. 4,010,116. The aluminosilicates of this latter
patent have high silica (SiO2) to alumina (A1203) molar
ratios, preferab1y above 4, to give high thermal stability~
RI -4078C
_35_ ~4478~
The fo]lowing is an example of a zeolite produced by the
silication of clay. A reaction composition is produced from a
mixture of sodium silicate, sodium hydroxide, and sodium chloride
formulated to contain 5.27 mole percent SiO2, 3.5 mole percent
Na20, 1.7 mole percent ch]orine and the balance water. 12.5 parts
of this solution are mixed with 1 part by weight of calcined kaolin
clay. The reaction mixture is held at about 16C (60DF) to 24C
(75F) for a period of about four days. After this low temperature
digestion step, the mixture is heated with live steam to about 88C
(190F) until crystallization of the material is complete, for example,
about 72 hours. The crystalline mater~al is filtered and washed to
give a silicated clay zeolite having a silica to alumina ratio of about
4.3 and containing about 13.5 percent by weight of Na20 on a
volatile free basis. Variation of the components and of the times
and temperatures, as is usual in commercial operations, will produce
zeolites having silica to alumina mole ratios varying from about 4 to
about 5. Mole ratios above 5 may be obtained by increasing the
amount of SiO2 in the reaction mixture. The sodium form of the
zeolite is then exchanged with polyvalant cations to reduce the
Na20 content to less than about 5 percent by weight, and
preferably less than 1.0 percent by weight. Procedures for
removing alkali metals and putting the zeolite in the proper form
are well-known in the art as described in U. S . Patent Nos.
3,293,192; 3,402,996; 3,446,727; 3,449,û70; and 3,537,816.
As previously indicated, the carbon black ingredient o~ the
present invention may be admixed with the zeolite forming
composition prior to crystallization of the zeolite and the carbon
black component later burned out to form large feeder pores
between the zeolite crystals in the resulting agglomerate.
Rl -4078C
~$~
-36- ~Z4~7~39
The zeo]ites and/or other promoters can be suitably dispersed
in matrix materials for use as cracking catalysts by methods
well-known in the art, such as those disc]osed, for example, in
U . S . Patent Nos . 3,140,299 and 3,140,253 to Plank, et al; U . S .
Patent No. 3,660,279 to Blazek, et al; U.S. Patent No. 4,010,116 to
Secor, et al; U.S. Patent No. 3,994,982 to Mitchell, et al; and U.S.
Patent No. 9,079,019 to Scherzer, et al.
The amount of zeolitic material dispersed in the matrix based
on the final fired product should be at least about 10 weight
percent, preferably in the range of about 2S to 40 weight percent,
most preferably about 35 to 40 weight percent. More than one type
of zeolitic particles, such as zeolites with different functions and/or
selectivities, and zeolitic particles in combination with parbcles of
metals or other catalytic materials may be emp]oyed together to make
up these amount of total promoter ingredient. For example,
particles of hydrodesulfurizing catalyst as in U. S. Patent No.
3,770,617 n~ay be mixed with particles of cracking catalyst as in
4,010,116. The upper ranges of zeolite concentrations, together
with a much less but significantly active matrix, are preferred to
provide a catalyst with ultrahigh cracking activity even at large
pore vo]umes and reiatively low surface areas.
Crystal]ine aluminosilicate zeolites exhibit acidic sites on both
interior and exterior surfaces with the largest proportion of total
surface area and cracking s~ies being internal to the particles
within the crystalline micropores. These zeolites are usually
crystallized as regularly shaped, discreet particles of approximately
0.1 to 10 microns in size and, accordingly, this is the size range
normally provided by commercial catalyst suppliers. Tc increase
3~ exterior (portal) surface area, the particle size of the zeolites for
the present ;nvent;on is preferably in the range of less than 0.1 to
RI-407~C
~`
7~39
-37--
about 1 micron and more preferably in the range of less than 0.1
micron. The preferred zeolites are thermally stabilized by heating
to temperatures of at least 538C (1000F) and then further
exchanged with hydrogen ions and/or rare earth ions. These
zeolites are steam stable to about 899C (1650F).
Feeder pore content of the catalyst within the size range of
about 400 to 6000A is provided in the main by the preparative
method of the invention. Macropores of less than 400A may be
produced by this me-thod and may also be introduced into the
catalyst matrix by prior art methods for preparing such matrices.
The micropore content of the ca-talyst in the size range below 20A is
provided principally by the zeolite particles per se.
Since the amount of zeolite used may vary widely, the
micropore content of the catalyst will be variable by a similar
amount. The fraction of total pore volume provided by micropores
of 20A or less also will vary depending upon the fraction of large
feeder pores introduced by the carbon black component.
The surface area of the zeolitic component in the catalyst may
be estimated by multiplying -the surface area of the pure zeolite
(usually about 800 to 900 m /gm) by the percentage of zeolite in
the final catalyst. Total pore volume including the feeder pore
volume of the matrix should be at least 0.2 cc/gm, preferably more
than 0 . 4 cc/gm . The upper limit for total matrix pore volume is
best expressed as the amount of contact surface area provided in
the final catalys-t matrix. In general, the minimum surface area of
a satisfactory catalyst matrix is about 20 square meters per gram,
preferably at least 30 square meters per gram, more preferably at
least 40 square meters per gram. These parameters are believed to
provide sufficient volume in the form of large feeder pores to
reasonably minimize diffusion limitation effects, yet not so large a
RI -4078C
~24~789
-38-
volume as to unduly reduce the availability of acidic sites or
adversely affect the physical properties, e.g., attrition resistance,
of the catalyst.
The introduction of CB feeder pores, as a practical matter,
causes no appreciable change in the volume and size distribution of
micropores in the zeolite and macropores produced in the matrix by
other mechanisms . Thus, the macropore conten-t of the ca talyst
particles produced by the carbon black of the invention is readily
obtained by the difference in pore volume between the unfired and
10 fired catalyst composition, provided there is no significant sin-tering
of the matrix material. The determination of feeder pore volume is
by the conventional mercury porosimetry method referenced above.
In order to test the activity of a catalyst, it is the practice in
the petroleum cracking art to measure the catalyst activity by a
bench-scale test. Although various tests have been proposed, one
test widely accepted by the industry at -the present time is known
as the microactivity test or "MAT". The microactivity test and
standard procedures for obtaining "MAT Activity" are described
below. Because the MAT test range is appropriate for equilibrium
20 (used) zeolite catalyst instead of virgin (unused) catalyst, virgin
catalyst of the present invention is treated with 97-100% steam at
787C (1450F) for 5 hours to provide a standard reduction in its
activity before it is tested by the MAT procedure.
Having thus described the catalyst of the invention broadly,
25 the following examples are offered to illustrate method for its
preparation in more detail.
Example 1
RI-4078C
3L~447&~9
-39 -
2.1 gms of carbon black pellets (N-339 from Ashland Chemical
Company) were placed in 350 ml of deionized water containing 0.03
gm of hexadecyltrimethylammonium bromide and O .03 gm of Triton
X-100 dispersants. This mixture was agitated ultrasonically with an
Artex Sonic probe at 300 watts for 30 minutes, and the resulting
carbon black suspension added to 700 ml of water along with 22.2
gm of Kaopaque 10 Kaolinte. This slurry was mixed for 30 minutes
with a Premier High Viscosity Dispersator and then 23.4 gm of rare
earth zeolite was added, along with 5 ml of PQ N-Brand sodium
metasilicate as a dispersant for the clay and zeolite. The resulting
slurry was mixed for an additional 30 minutes with the Dispersator
and then 120 ml of sodium metasilicate and 200 ml of water were
added with mixing for 10 more minutes . 180 ml of a 11.5 wt%
sulfuric acid solution were added slowly while mixing continued to
partially neutralize the slurry and precipitate silica gel. This silica
gel contained uniformly dispersed carbon black and clay and was
aged for 1 hour at 43C (110F) to increase the size of interstitial
cavities within the gel. Mixing was then resumed and a 18 wt%
solution of aluminum sulfate, prepared from 108.4 gm
Al2(S04)3:18H20 and 200 ml of water at 49C (120F), was added to
the silica gel and mixing continued for 15 minutes at 43C (110F).
This slurry was then neutralized by adding 68 ml of concentrated
NH40~I to precipitate alumina gel. The silica and alumina gel
mixture was filtered and the resulting gel cake washed three times
with 3L of water at 66C (150F). The washed gel cake was then
dried at 260C (500F) for 16 hours and the dried cake ground into
small solid particles. These particles were then exchanged at ~3C
(200F~ for 1 hour with lL of 1. OM rare earth chloride solution .
The exchanged particles were filtered and washed three time wi th
2L of water at 6~C (150F) and dried a-t 260C (500F) for 16
hours. After drying, the carbon black was burned out by firing
the catalyst particles in air at 538C (1000F) until the catalyst
became white in color (about 3 hours). This procedure produced
RI -4078C
~2~L47~39
-40-
about 100 gm of dry catalyst particles consisting essentially of 20
wt% zeolite and 20 wt% clay substantially uniformly dispersed in 60
wt% of a silica-alumina matrix containing about 75wt~o SiO2.
Examples 2-5
-
S The quan$ity of carbon black used in Example 1 above
provided about 2 wt% carbon black in the unfired catalyst
composite. In l~xamples 2-S, the same preparation procedure was
followed but the amount of carbon black was changed to 4.â, 11.6,
27 .1 and 44 . 6 gms giving a carbon black wt% in the unfired
composites oE 4, 10, 20, and 30, respectively. In these examples,
the amounts of the other ingredients remained the same except that
the amount of carbon black dispersant was varied in proportion to
the amount of black used. Thus, the weights of
hexadecyltrimethlyammonium bromide and Triton ~-100 used in these
examples were each approximately 1% of the weight of carbon black
added to the deionized water.
The changes in pore volume produced by the carbon black of
Examples 1 through 5 are shown in Table 1 relative to a fired
control sample containing no carbon black. In order to show the
effects of firing alone on pore size distribution the control sample
without carbon black was also fired and the resulting pore volume
changes for this sample are reflected in the first column of Table 1
under the heading "Control".
The data shown in Table 1 indicates that firing of the control
sample containing no carbon black caused a slight decrease in pore
volume in all pore size ranges. By comparison, the firing of
s?r~ples containing carbon black resulting in significant increases of
porosity in the ranges abo~e about 400A and 1000A. Several
composites also showed significant increases in porosity near 60A
RI -4078C
:~LZ~7~39
-41 -
for reasons not yet explained, but which could involve differences
in aging the hydrogel. Sample No. 5, which contained 30 wt%
carbon black, did not form an adhering cake when filtered and
dried, but instead produced a fine powder as an unfired product.
RI -4078C
~478~
O ~OO~
O ~ ~1 0 ~ O O ~ ~ ~ ~
ooooooo~ ~ ~ oo
~a ooooooOoo o o o
U~ ~
o ,~
~ ~ ~r~co~o~
E~ ~ o ,/ o ~ ~ o ~ ,~ n o
O O O ~1 0 0 0 0 0
,
ooooooooo o o o
~ l l
~4 ,~
o n o ~ ~ ~J oo ~ ~ n u~ I~
O ~ ~ ~ O O ~ ~ ~ ~ I~ CO 1
~ O ~ OOOOOOOOO ~1 ~ `J
Z ~ ........ ~
o U~ ooooooooo o o o
t, ~ ~
~ ~ I~ln~n o ~ o
O ~ ~1 ~ ~ ~ O ~1 ~ ~1
oo,,oooooo o
u~ R .........
E~ ,~ ooooooooo o o o
~ O
U~ ,~
V S~
~0
~1 R ~
I~l~`J~o a~ 1~ `J
~J ~0~0000~ ~1 ~ 1
~ ~ ~~ oo~oooooo o ~ I~
H O . .... .. ...
U~ ~r~ OOOOOOOOO O O O
p; ~ 3
O
O ~ a~ ~ ~ 1~ o u~ ~ O ~ ~ a~ c~
,1 ~ o ~ o ~ ,1 o ,1 o ~ C~ U~
~- o~ ooooooooo o o
,~ ~::h E~ .........
o O~ ooooooooo o o o
~ t~ ~IIIIIIII I I
~q
~ ..
o¢
a
.,,
a)
~o
3~ ~¢
O~ O ~ AO ~
ooC.)O O~ O
o o o o o
0 0 0 0
u~ o o o o c~
o ,1 a ~ , 0 ~ 0
Z ~4 ~ O U~ O O O O O O O O ,~ ~ ~
a ~ ~ 4 a ~o
,1 o h ~ ,I ~d
ra o a~ o
h ~ ~
t~ ~o o o o O
t~ ~ P~
.
-43- ~;~44'7~39
Examples 6-9
The catalyst samples of Examples 6 throug 9 were prepared
following the same basic procedure of Examples 1 through 5 except
the amount of clay was increased and the amount of silica and
alumina gel decreased so that the amounts of clay and of
silica-alumina gel were approximately equal in the final fired
product, wlth the resulting silica-alumina matrix containing about 75
wt% SiO2. In addition, instead of filtering the drying to produce a
gel cake, the final neutralized slurry was spray dried -to Eorm
microspheres of catalys t . These microspheres were then washed,
exchanged with rare earth chloride solu-tion, rewashed, and dried at
149C ~300F) for 16 hours in a manner similar to the gel cake and
ground particles of Examples 1-5. Spray drying produces smaller
and more unifor m particle sizes than grinding gel cake and is
preferred for providing a more niform and fluidizable catalyst. The
average size of the spray dried particles is preferably within the
range of about 40 to 80 microns.
The changes in pore volume produced by the carbon black of
Examples 6 through 9 are given in Table 2. In this comparison,
the changes in pore voume were measured relative to a fired control
sample containing no carbon black instead of the unfired type of
control sample forming the basis for the comparison in Table 1.
The changes in porosity exhibited by the samples of Table 2 are
consistent with those of Table 1 in that the use of carbon black
increased catalyst porosity in the ranges above 400A and 1000A.
Howevar, no significant increase in porosity occurred in the 40 to
80A range, the reason for this reason for this variance with the
data Table 1 being uncertain at the present time.
RI -4078C
~2~
-44 -
TABLE 2 - POROSITY EFFECTS OF CARBON BLACK ADDITIONS
Sample No. - 6 7 8 9
Carbon Black,wt% - 11.6 22.0 23.4 27.0
Pore Volume Change in cc/gm as compared to
fired control sample without carbon black.
Pore Diame~,er
Range in A
30-35 ~ .004 0.022 0.017~0.006
10 35-40 -0.001 0.004 0.001 0.001
40-60 - .007 -0.051 -0.042 -0.068
60-80 - .003 -0.035 -0.026 0.042
80-100 0.006 -0.009 -0.013 -0.011
100-200 0.020 -0.005 -0.003 -0.010
15 200-400 0.027 0.009 0.008 0.004
400-1000 0.058 0.044 0.022 0.014
1000-6000 0.233 -0.032 0.481 0.225
>6000 * 0.233 -0.032 0.481 0.225
Group B
20 1000->6000 * 0.342 0.201 0.708 0.450
400->6000 * 0.400 0.245 0.730 0.464
200- >6000 * 0.427 0.254 0.738 0.468
100->6000 * 0.447 0.249 0.735 0.458
Group C
25 400-6000 0.167 0.277 0.249 0.239
200-6000 0.194 0.286 0.257 0.249
100-6000 0.214 0.281 0.251 0.233
Group D
200-1000 0.085 0.053 0.030 0.018
30 100-1000 0.105 0.048 0.027 0.088
* Includes pore sizes above 6000A.
RI-4078C
~z~
-45-
In Figure 6 of the drawings, the changes in the pore volume
are plotted against carbon black content for several of the pore
diameter ranges in Table 2. The curves of Figure 6 correspond to
the data of Table 2 as follows: curve A-7 represents the pore
5 volume change in the 7th pore diameter range of group A; curve
A-8 represents the pore volume change in -the 8th pore diameter
range of group A; curve D-2 represents the pore volume change in
-the second pore diameter range of group D; and curve C-3
represents the pore volume change in the third pore diameter range
10 of group C.
The curves of Figure 6 suggest that the maximum number of
feeder pores having diameters in the range of about 100 to 6000A
occur with 18 to 19 wt% carbon black in the unfired catalyst
composite. This amount of carbon black, which equals about 50% by
15 weight of the dry silica-alumina gel in the unfired composite,
increases the volume of pores having diameters greater than 100A
by about O . 3 cc per gm . Similarly, a carbon black concentration of
about 11 wt% in the unfired composite, which is about 30% by weight
of the silica-alumina gel present, provides the maximum number of
20 feeder pores in the lQO to 1000A range. This corresponds to an
increase in pore volume of about O.1 cc per gm. As also evident
from Table 2, most of the porosity increase at 12 wt% carbon black
is due to feeder pores in the 400 to 1000A range. Those feeder
pores increased the pore volume by about O.07 cc/gm, which is
25 about the volume of the carbon black burned out of the sample.
The maximum porosity increase in the 200 to ~OOA pore diameter
range occurred at about 10 wt% carbon black.
The curves of Figure 6 further indicate that there is an
optimum carbon black concentration to attain maximum porosity in a
30 given pore size range and that lower concentrations are rec~uired to
give the maximum porosity for smaller diameter pores. This
RI -'1078C
2447~3g
-46-
suggests that the pore sizing attainable with carbon black is closely
re]ated to the deyree to which individual reticu]ated carbon black
particles are separated from each other and dispersed within the
silica-alumina matrix, full dispersion becoming more difficult to
attain and maintain as concentrations of carbon black in the gel
slurry increase. In addition to using lower concentrations of
carbon black, more effective carbon black dispersants also increase
the proportion of pore sizes in the sma]ler diameter ranges of
feeder pores. One example of a more effective dispersant for
10 carbon black is sodium lignosulfonate which may be used in place of
hexadecyltrimethlalTunonium bromide and/or Triton X-100
Aispersants, particularly where more feeder pore diameters
approximating the diameter of primary carbon black particles are
desired .
Lignin is derived from wood pulp and varies in molecular
weight be~ween 1000 and 50,000. It's basic organic structural unit
is a substituted phenylpropane. Lignin dispersants are
commercially available under the name Marasperse which is a sodium
lignosulfonate of relatively low sulfonation. While not intending to
20 be bound by any one theory or hypothesis, the action of li~nin as
a dispersant is believed to be electrochemical in nature. When
lignin molecules are adsorbed on the solid carbon black or clay
particles in aqueous suspension, they impart a negative charge to
the particles, causing them to repel each other. Adsorption of
25 lignin molecules on a particle may also create a film which then acts
as a physical barrier against direct contact between the particle and
the surrounding aqueous media, inc]uding silica colloids. These
effects are believed to contribute to the production of feeder pore
sizes corresponding generally to the transverse diameters of
30 primary reticulated carbon black particles and to improved pore size
control. Sodium lignisulfonate is a preferred dispersant for carbon
black in the catalyst composition of the invention and is added to
*Trademarks
RI-~078C
~'
7~9
-47-
the matrix slurry with the carbon b]ack prior to the spra~ drying
or other forming step for shaping the so]id cata]yst particles. The
cata]yst compositions containing ]ignin and the processes for making
those compositions are attributed to William P. Hettinger, Jr., James
E. Lewis and H. Wayne Beck, a]l of Ash]and Oil, Inc.
Examples 10-11
In a Kady Mill*homogenizer, 12 liters of tap water was mixed
with 960 grams of aluminum sulfate hydrate and 1.2 liters of H2SO~.
Using an addition rate of 200 ml/min., 9 liters of "N" Brand sodium
metasilicate was added to this acid solution while mixing to provide
a silica hydrosol. Ten grams of a lignin dispersant called
marasperse and 800 grams of N-219 carbon black were added to the
homogenizer ancl mixed for 5 minutes to disperse carbon black in
the silica hydrosol. In a separate container, 10 grams of sodium
pyrophosphate (Na4P207) was added to 11 liters of 2pH water made
with H2SO4. With vigorous mixing, 11 kg of fine kaolinite clay was
added to form a clay slurry. The clay slurry was then combined
with the silica hydrosol in the homogenizer and mixed for 5
minutes .
A slurry of NaY ~:eolite made from 4 liters of 2p~ H20 and 4
kg of zeolite was quickly added to the homogenizer and mixed for
10 minutes. The resulting slurry was immediately spray dried at
400C inlet and 125C outlet temperatures in a Niro Atomizer Model
V*Spray Drier to form catalyst microspheres. Air pressure was 30
psig. One kilogram of the microspheres from the spray drier was
washed three times with 4 liters of 66C (150F) water filtered.
The filter cake was exchanged twice with 3 liters of 1. 25 Molar
NH3Cl for 15 minutes each at 66C ~150F~. Alter filtering again,
the cake was exchanged 3 times at 66C (150F~ for 30 minutes each
with 3 liters of 0 . 33N mixed rare earth ch]oride so]ution . The solid
*Trademarks
RI-4078C
~.
-
~;Z4~7~9
-48-
particles were then finally washed 4 times with 3 liters of 66DC
(150F) water and dried at 149C (300F) for 16 hours to produce
Sample no. 10. A portion of this sample was oxidized in air at
538C (1000F) in a shallow ~ed for 4 hours to burn out the carbon
5 black and produce Sample no. 10R.
Samples 11 and 11R were prepared by the same procedure
except twice as much carbon black was used, the designation "R"
meaning regenerated (fired) to burn out the carbon black particles.
In the preferred embodiment of ehe invention, the fired
10 microspheres are -further exchanged a~ter firing wi-th a 0.5N
solution of mi~ced rare earth chlorides at 66C (150F) for 30
minutes, and then washed 4 times with 66C (150F) water and
dried for 16 hours at 149C (300F). The firing step to remove
carbon black also calcines the zeolite in the composition This
15 calcination in combination with the fur-ther ion exchange provides an
ultrastable, hydrothermally resistant zeolite catalys t ~ A portion of
Sample 11R was further exchanged in this manner to produce
Sample 11RE shown in Table 5.
Table 5 gives some of the catalytic properties of Samples 10R,
20 11R and 11RE as compared to the Davison catalyst. The data given
was determined by a micro-activity test (MAT) based on the
procedure found in ASTM test method no. D-3908-80 which is
described in more detail below. Although an FCC type feed is used
in this standard test so that it is not fully indicative of the
25 performance of the catalyst samples with a carbo-metallic feed, the
data is believed to show the advantage of the invention in reducing
the carbon make. Thus, the carbon producing factor (CPF as
defined below) and the weight percent coke (rela-tive to feed
weight) are significantly lower for Samples 10R, 11R and 11RE of
30 the inven-tion relative to the Davison catalyst as seen in Table 5.
RI -4078C
Z4~9
-49-
It is also significant that the MAT conversion obtained with Sample
10R was 80 as compared to 73 obtained with the Davison catalyst.
Referriny to Table 3, it will be seen that the higher conversion was
obtained with less zeolite, namely, 6.6 percent intensity for Sample
5 10R relative to 13.1 percent intensity for Super DX. The relatively
high conversion and low coke producing factors of Sample 10R
demonstrate clearly that the heavier hydrocrabons of even a MAT
feed can be cracked by and stripped more effec-tively from the
catalyst of the present invention. The lower conversion
10 experienced with Samples 11R and 11RE are attributed to the very
low level of zeolite measured for those samples, namely, 3.8 percent
intensity .
RI-4078C
~447~3~
-50-
Table 3 - Composition of Samples 10R and 11R
Compared to Davision Super DX Catalyst
Davison
Sample 10RSample 11R Super DX
5Amount of Carbon
13lack Burned
Out, wt% (1) 4.5 8.3 None
Al203, wt% 29 . 6 27 . 0 28 . 3
SiO3, wt% 63 . 2 55 . 3 64 . 2
10Na20, wt% 0 . 33 0 . 22 0 . 84
Zeolite Intensity,
NaY (2~ 6.6 3.8 13.1
Hg (<6000A) Pore
Volume, cc/gm 0 . 25 0 . 33 0 . 22
15Surface Area, m2/gm 198 197 197
(1) Carbon Black wt% based on unfired Samples 10 and 11. All
other wt% are on ignition basis (IB) of final fired catalyst.
(2) Zeolite X-ray intensity of sample as percentage of X-ray
intensity of high purity sodium Y zeolite.
Rl-4078C
~1 ~2447~39
oo O r~ o ~ o o ~ u~
U
~1 ~ o~ ~ O r~ o u~
oo e ~ D o ~7 0 ~ cO 00
_ O O ~ O O O O U~ U~
~o~ oooooooooo
U o o o o o o o o O O
,~ ~ ~ o ~ I~ ~ ~ r~
1 0~ 0~ o
U~ ~ O O O O O O O O
U
U o O O O O O O O O O
-
~,
~,
_I ~ ~' ~ o O
~ OOOOOOOO~
a ~ "
:~
aJ
~a
~n
X ~,
o~ ~ ~ ~ o o o 1` o ~ ~
U . . . . . . . . . . ~,
U~ ~ O ~ u~ ~ o ou~ r~ o ~o~
o _I
P~ o
,.
,~ ~
", e~ ^
o ~ o In ~C~ --~ O ~1oO
~ ~ ~l ~ o o o o ~u~ -
E~O~u OoooooooOo
t~ u o o o o o o o o o o
~ o
c`
~_ a
~0 00a~ ~ O ~,0,~ ~~0
~ ~ o o o o o o o o ~c~ ~
u~ ~o o o o o o o o oo s~
~o
-~ ~u p~
~l ~l `r
o ~
~ e G~ ~ ~ o
~ oo ~ ~ ~ ~ oo ~ ~ ~ C~ s~
u~ ~ o o o o oo o o ~ ~ os~ o
u o o o o oo o o o o ~~ ~
d
. ,_ o ~~ U
o o o
, O
a~ ,, ~ C' co o ~D ~
ooIII~I'C
oOOOoIIIOE~
o K ~ ~ ~ ~ ~ ~ '' '`~
~Z4~789
-52-
Table 5 - Catalytic Properties After
Steaming (1) of Samples 10R, 11R and 11RE
Compared to Davison Super DX Catalyst
Sample Sample Sample Super
lOR 11R 11RE DX
MAT Conversion,
Vol% 80 59 . 7 58 . 3 73
HPF 1.3 2.1 0.7 0.6
CPF 0.7 0.8 0.5 1.1
Coke, wt% Feed 2.6 1.3 0.7 2.8
(1) Each sample was treated with steam (97%) for 5 hours at 787C
(1450~F) to simulate a used catalys-t at equilibrium FCC
conditions .
For comparison purposes with reference to the data of Table 5,
two additional samples, designated as Sample Nos. 12 and 13, were
prepared following the preparation procedure described above for
Sample No. 10, but without the addition of carbon black. These
samples yielded MAT volume percent conversions of 73. 5 and 79.7,
respectively, and carbon producing factors (CPF) of 0.92 ~nd 0.84,
respectively, with zeolite intensities of 7.3 and 12 . 5, respectively .
The catalyst of the invention has a relatively high level of
cracking activity and is capable of providing high levels of
conversion and selectivity at low residence times in the riser. The
conversion capabilities of the catalyst may be expressed in terms of
the conversion produced during actual operating of an RCC
cracking or other conversion process and/or in terms of conversion
produced in standard catalyst activity tests.
For example, it is preferred to employ a catalyst which, in the
course of ex-tended operation under prevailing process conditions, is
30 sufficiently active for sustaining a level of conversion of at least
about 50 percent and more preferably at least about 60 percent. In
thiS connection, conversion is expressed as liquid volume percent
based on volume of fresh feed. Conversion is the volume
RI-4078C
789
-53-
percentage of feedstock that is converted to 221C (430F) endpoint
gasoline, lighter products and coke, and is calculated by
subtracting from 100 the volume percentage of those products
heavier than the gasoline which remain in the recovered product.
Also, for example, the preferred catalys t may be defined as
one which, in its virgin or equilibrium state, exhibits a specified
activity expressed as a percentage in terms of MAT (micro-activity
test) conversion. For purposes of the present invention, the
foregoing percentage is the volume percentage of standard feedstock
which a catalyst under evaluation will conver-t to 221C (430F) end
poin-t gasoline, lighter products and coke at 482C (900F), 16
WHSV (weight hourly space velocity), and 3 C/O (catalyst to oil
weight ratio) using the equipment and procedures specified by
ASTM D-32 MAT test D-3908-80 and an appropriate standard FCC
feedstock. The WHSV is calculated on a moisture free basis using
clean catalyst which has been dried at 593C (1100F), weighed and
then conditioned for a period of at least 8 hours at about 25C and
50% relative humidity, until about one hour or less prior to
contacting the feed. The feedstock is preferably a sweet light
primary gas oil, such as that used by the Davison Divison of W.R.
Grace and defined as follows:
API Gravity at 16C (60F), degrees 31.0
Specific Gravity at 16C (60F), g~cc 0.8708
Ramsbottom Carbon, wt% 0.09
Conradson Carbon, wt% (est . ) 0.04
Carbon, wt% 84.92
Hydrogen, wt% 12.94
Sulfur, wt% 0.68
Nitrogen, ppm 305
Viscosity at 38C (100F), centistokes 10.36
Watson K Factor 11.93
Aniline Point 182
Bromine No . 2.2
Paraffins, Vol% 31.7
Olefins, Vol% 1.8
RI-4078C
4~
-54-
Naph thenes, Vol% 44 . 0
Aromatics, Vol% 22 ~ 7
Average Molecular Weight 284
Nickel Trace
Vanadium Trace
Iron Trace
Sodium Trace
Chlorides Trace
B S & W - Trace
10 Distillation ASTM D-1160
IBP
10% 601
664
50% 701
15 70% 734
90% 787
FBP 834
The gasoline end point and the boiling temperature volume
percentage relationships of the products produced in the MAT
20 conversion test may be determined by simulated distilla tion
techniques, for example by modification of the gas chromatographic
"Sim-D" technique of ASTM D-2887-73. The results of such
simulations are in reasonable agreement with the results obtained by
subjecting larger samples of material to standard laboratory
25 distillation techniques.
The catalyst may be introduced into the process of the
invention in its virgin form or in other than its virgin form, e.g.,
one may use equilibrium catalyst withdrawn from anotller unit such
as catalyst that has been employed in the cracking of an ~CC ~eed.
30 Whether characterized on the basis of MAT activity or relative
activity, the preferred catalysts may be described on the basis of
their activity "as introduced" into the process of the present
invention, or on the basis of their "as withdrawn" or equilibrium
activity in the process of the present invention, or on both of
35 these bases.
RI-4078C
~. .
Z44~
-55-
A preferred activity level of virgin and non-virgin catalyst "as
introduced" into the process of the presen-t invention is at least
about 50%, preferably at least about 60% by MAT conversion.
However, it will be appreciated that, particularly in the case of
5 non-virgin catalys-ts supplied at high addition rates, lower activity
levels may be acceptable. An acceptable "as withdrawn" or
equilibrium activity level of catalyst which has been used in the
process of -the present invention is at least abou t 50% and an
activity level of 60% or more on a MAT conversion basis is also
10 contemplated. More preferably, it is desired to employ a catalys-t
which will, under -the conditions of use in the unit, establish an
equilibrium activity at or above the indicated levels. Catalyst
activities are determined with the catalyst having less than 0. 01
coke, i.e., fully regenerated catalyst.
In Table 5, "CPF" stands for Carbon Producing Factor and is
defined as the ratio of the amount of coke produced by the test
catalyst to the amount of coke produced by a standard catalyst at
the same conversion level. "HPF" stands for Hydrogen Producing
Factor and is defined as the ratio of the amount of hydrogen
20 produced by a standard catalyst at the same conversion level. The
standard catalyst is chosen from amonç~ conventional FCC catalysts,
such as for example, zeolite fluid cracking catalysts, and is chosen
for its ability to produce a predetermined level of conversion in a
standard feed under the conditions of temperature, W~ISV (weight
25 hourly space veloci~y), catalyst to oil ratio and other conditions set
forth in the preceding description of the MAT conversion test and
in ASTM D-32 MAT test D-3907-80. For standard feed, one may
employ the above-mentioned light primary gas oil, or an equivalent
FCC feed. Although the equipment employed for the referenced
30 MAT test is not capable of processing RCC feeds, applicants are
under-taking to develop an equivalent tes t for evaluating RCC
catalysts using RCC type feeds.
RI-4078C
` -- -
~LZ4~7~
-56-
On pages 935-937 of Hougen and Watson, "Chemical Process
Principles" , John Wiley & Sons , Inc ., N . Y . (1947) , the concept of
"Activity Factors" is discussed. This concept leads to the use of
"relative activity" to compare the effectiveness of an operating
catalyst against a standard catalyst. Relative activity measurements
facilitate recognition of how the quantity requirements of various
catalysts differ from one another. Thus, relative activity is a ratio
obtained by dividing the weight of a standard or reference catalyst
which is or would be required to produce a given level of
conversion, as compared to the weight of an operating catalyst
(whether proposed or actually used) which is or would be required
to produce the same level of conversion in the same or equivalent
feedstock under the same or equivalent conditions. Said ratio of
catalyst weights may be expressed as a numerical ratio, but
preferably is converted to a percentage basis.
For purposes of conducting relative activity determinations,
one may prepare a "standard catalyst curve", a chart or graph of
conversion (as above defined) vs. reciprocal WHSV for the standard
catalyst and feedstock. A sufficient number of runs is made under
ASTM D-3907-80 conditions (as modified above) using standard
feedstock at varying levels of WHSV to prepare an accurate "curve"
of conversion vs. WHSV for the standard feedstock. This curve
should transverse all or substantially all of the various levels of
conversion including the range of conversion within which it is
expected that the operating catalyst will be tested. From this
curve, one may establish a standard WHSV for test comparisons and
a standard value of reciprocal WHSV corresponding to that level of
conversion which has been chosen to represent 100% relative activity
in the standard catalyst. For purposes of the present disclosure,
~0 the aforementioned reciprocal WHSV and level of conversion are,
respectively, 0 . 0625 and 75%. In testing an operating catalyst of
unknown relative activity, one conducts a sufEicient number of runs
RI -4078C
-57-
with that catalyst under D-3907-80 conditions ~as modified above) to
establish the level of conversion which is or would be produced
with the operating catalyst at standard reciprocal WHSV.
Then, using the above-mentioned standard catalyst curve, one
5 establishes a hypothetical reciprocal WHSV constituting the
reciprocal WHSV which would have been required, using the
standard catalyst, to obtain the same level of conversion which was
or would be exhibited by the operating catalyst at standard WHSV.
The relative activity may then be calculated by dividing the
lG hypothetical reciprocal WHSV of the s-tandard catalyst by the actual
reciprocal Wl~ISV of -the test catalyst. The resul-t is relative activity
expressed in terms of a decimal fraction, which may then be
multiplied by 100 to convert to % relative activity. Relative Activity
may also be expressed as follows: relative activity at constant
15 conversion is equal to the ratio of the WHSV of the test catalyst
divided by the WH~V of the standard catalyst. To simplify this
calculation, a MAT conversion vs. relative activity curve was
developed utilizing a standard catalyst of 75 vol% conversion to
represent 100% relative activity. One such curve is shown in
20 Figure 14. In applying the results of this de-termination, a relative
activity of 0. 5, or 50%, means that it would take twice the amount
of the operating or test catalyst to give the same conversion as the
standard catalyst , i . e ., the production catalyst is 50% as active as
the reference catalyst.
The relative activity level of virgin and non-virgin ca-talyst "as
introduced" into the process of the present invention is at least
a~out 20%, preferably at least about 40% and more preferably at
least about 60%. However, it will be appreciated that, particularly
;n the case of non-virgin catalysts supplied at high addition rates,
lower activity levels may be acceptable. An acceptable "as
withdrawn" or equilibrium relative activity level of catalyst which
XI-4078C
Z~47~3~
-58-
has been used in the process of the present inven tion is at least
about 20% or more, preferably about 40% or more, and more
preferably about 60% or more. More preferably, it is desired to
employ a catalyst which will, under the conditions of use in the
5 unit, establish an equilibrium activity at or above the indicated
levels .
Referring to the idealized carbon particle illustra-ted in Figure
1, -the individual reticulated CB particle "R" is made up of a series
of primary particles "P" of roughly spherical shape held -together
10 by atomic and/or interfacial binding forces. At-tached to a central
chain C-C are a number of side chains S1, S2 and S3, the actual
number of side chains varying from particle to particle and the
extent of such branching being indicated by a "branching factor",
such as given in Figure 9B through 12B. The diameter of the
15 particle P determines the transverse diameter of the central chain
C-C and is usually about 100 to lOOOA, preferably about 300 to
600A. The length of the longest chain, such as from point L to
point M in Figure 1, is usually about 500 to 30,000A, preferably
1000 to 10, OOOA . The average length to diameter ratio for such
20 particles is preferably in the range of 2 to 5 or greater. It is
believed that such reticulated particles yield feeder pores roughly
of the same dimensions as illustrated by the dotted outline of a
feeder pore F in Figure 1.
The central chain and these side chains are believed to overlap
25 with those of adjacent reticulated particles and thereby provide
correspondingly tortuous channels and feeder pore branches in the
matrix leading to the zeolite particles supported therein. In
addition, two or more reticulated particles may adhere closely
together to form an agglomerate mass so that the resulting pore size
30 corresponds to the dimensions of this mass. Such agglomerates are
believed to produce pore sizes of lQOOA and greater. The degree
RI-4078C
:
3 2~ 9
-59-
of agglomeration depends not only upon the particular dispersant
used, but also upon the mixing means used and the mixing energy
imparted to the carbon black suspension and resulting catalyst
slurry. A greater percentage of pore sizes in the range o~ 400 to
5 600A may be realized with catalyst compositions similar to those of
Examples 1 to 9 but using increased agitation in combination with
liynosulfonate as -the dispersant.
A scanning electron microscope (SEM) was used to make
micrographs of some of the carbon blaclc particles useful in the
l() catalyst of the present invention. Copies of those micrographs are
presented in Fig~lres 9A, 10A, 11A and 12A. Contour plots were
also prepared of these reticulated carbon black particles by plotting
the particle contours with an X-Y recorder attached to a computer
read-out from photogrammetric topography. The contour plots,
15 together with tables of topographical measurements for each
particle, are presented in Figures 9B, 10B, 11B and 12B which
correspond to carbon black types N-339, N-550, N-220 and N-326
shown in Figures 9A-12A, respectively.
Figure 13 is a photograph showing a catalyst matrix from
20 which the carbon black has been removed, the catalyst composition
having been made and fired in accordance with the teaching of the
invention. The photograph has been marked with horizontal
coordinates 1 through 10 and vertical coordinates A-H. The dark
areas of the photograph are pores and the horizontal and vertical
25 coordinates may be used to identify the location of those pores.
The location and approximate size of the minor transverse dimension
of some of the carbon black pores are identified as follows: D1-2,
- O O O O O
300A, G4, 1500A, B6-7, 250A, C7-8, 700A; D9, 600A. As shown in
the upper left hand corner of Figure 13, the scale oE this
30 photograph is in micrometers (microns), one-tenth of a micrometer
being equal to 1000 Angstroms uni ts . The dimensional units of
RI-4078C
~2~7~
-60--
Figure 13 are comparable to those of Figures 9 through 12 where
the scale is in nanometers (nm), one nanometer being equal to ten
Angstrom units.
THE CONVERSION APPARATUS
Al though the above catalyst may be used in various types of
contacting operations within the contemplation of this invention, the
catalyst is particularly useful in the catalytic cracking of residuum
or carbo-metallic oil feeds. The catalytic cracking operation may be
conducted in various types of reactors and associated equipment,
such as fixed bed systems or fluidized systems. The ca-talyst is
par-ticularly useful in a fluidized bed type of operation where the
catalyst in a finely divided fluidized state is suspended in vaporized
feedstock and the suspension of catalyst and feedstock is passed
upward through an elongated riser in a progressive flow
arrangement without significant backmixing. In general, riser
operations are carried out at conditions conducive to achieving the
improved results desired, bearing in mind the specific feedstock,
catalyst composition and process equipment being used. The
process is preferably carried out without added hydrogen.
Figure 7 is a schematic diagram of one apparatus for carrying
out the process of the present invention. The carbo-metallic oil
feed is supplied through a feed supply pipe 10 having a control
valve -l1 to a wye 12 in which -the feed is mixed with a flow of
catalyst delivered through a catalyst supply pipe 13 as controlled
by a valve 14. When used, water or some other diluent may be
introduced to the wye through a diluen-t supply pipe 9. The
mixture of catalyst and feed, with or without additional diluent
materials, then flows upward through a riser reactor 18. The riser
18 is an elongated conduit in which the length to diameter ratio may
3~ vary widely. The relatively narrow riser conduit provides high
RI-4078C
.Z4~7~3~
-61-
linear velocities, such as in the range of 25 to 90 feet per second.
The length to diameter ratio of the riser is preferably in the range
of 10 to 30, more preferably 20 to 25.
At the upper end of riser 18 is a chamber 19 which receives
S catalyst from the riser. While chamber 19 may be a conventional
disengagement and collection chamber, often referred to as a
reactor vessel, it is preferred that means be provided at the riser
exit for causing product vapors to undergo a rapid change of
direction relative to the direction traveled by the catalyst particles
10 so that the vapors are suddenly or "ballistically" separated from the
catalyst particles. Within disengagement chamber 1~ of Figure 7 is
an upward extension 20 of riser pipe 18 having an open top 21
through which the catalyst particles are discharged. The product
vapors are caused to undergo a sudden change of direction into
15 lateral port 22 in the side of riser extension 20, the catalyst
particles because of their momentum being, for the most part,
unable to follow the product vapors into port 22. This embodiment
of the cracking apparatus makes use of the vented riser concept
described in U.S. Patent Nos. 4,066,533 and 4,070,159 to Myers, et
20 al. The
vented riser thus provides a substantially instantaneous separation
between hydrocarbon and catalyst in disengaging vessel 19.
Because of the relatively high severity required to crack the more
refractory components of carbo-metallic feeds, rapid disengagement
25 of catalyst from cracked hydrocarbons by ballistic separation
prevents overcracking of desirable liquid products, such as
gasoline, to light undesirable products, such as hydrogen and
methane gases and carbon.
The product vapors and a relatively small amount of catalyst
30 entrained with them are transferred by a cross pipe 23 to a c~clone
separator 2~ for removing the entrained catalyst. The cyclone
R~ -9078C
~2~47~
-62 -
separator 24 may be connected via a transfer pipe 17 to an optional
secondary cyclone separator 25. Product vapors are discharged
from disengagement chamber 19 through overhead product discharge
pipe 26.
The catalyst particles which discharge from open top 21 of
riser pipe extension 20, and those entrained catalyst particles which
are discharged from dip legs 27 and 28 of the primary and
secondary cyclonés 24 and 25, respectively, drop to the bottom of
disengagement chamber 19. Catalyst spilling over from the bottom
L0 of disengagement 19 passes via a drop leg 31 into a s tripping
chamber 32 equipped with baffles 33 and a steam jet 34. Other
stripping gases known in the art may be introduced through jet 34
and employed with or in place of steam.
Accumulated carbon is burned from the catalyst in a combustor
38 which receives stripped catalyst via a downcomer pipe 39 with a
control valve 40. Combustion air is supplied from an air supply
system, generally indicated by 44 and including blowers 41 and 42
and filter bank 43, to combustion air jets 48 at the bottom of the
combus-tor and to fluffing air jets 49 at an elevated position therein.
Regenerated catalyst, with most of the carbon burned off, departs
the combustor 38 through an upper outlet 50 and cross pipe 51 to a
secondary combustion chamber 52, where it is deflected into the
lower portion of chamber 52 by baffle 53. Although the use of
two-stage regeneration is contemplated and preferred, in this
particular embodiment, the secondary chamber 52 is operated
primarily as a separator chamber for separating regenerated catalyst
from combustion gases. Additional oxygen containing gas may be
introduced through lines 59 and secondary chamber 52 operated
with an excess of oxygen to insure removal of carbon deposits down
to residual carbon levels of 0.05 weight percent or less.
RI -4078C
~L2~47~39
--63-
Catalyst may optionally move in up to three different directions
from secondary chamber 52. Fully regenerated catalyst is
discharged through bottom outlet 69 to catalyst supply pipe 13
through which it is recirculated to wye 12 for contact with fresh
S feed as previously described. A portion of the hot catalyst may be
circulated back to combustor 38 via catalyst recirculation line 55
with control valve 56 for heat control in the combustor and to raise
the temperature of incoming air with hot catalyst so as to insure
that combustion of coked catalyst is properly initiated. Since some
of the catalyst will be entrained in the combustion gases, such as
carbon oxides produced by burning of the carbon, two sets of
primary and secondary cyclones, generally indicated by 57 and 58,
are provided in chamber 52 for separating these catalyst fines from
the combustion gases. Catalyst collected in cyclones 57, 58 is
discharged through their dip legs to the bottom of chamber 52
where the catalyst is kept in suspension by air, inert gas and/or
steam from lines 59 and by a baffle arrangement 54. Combus-tion
product gases produced by regeneration of the catalyst and
separated from entrained catalyst fines are discharged through
effluent pipes 61 62 and heat exchangers 60, 63. If such 0ases
con-tain sufficient amounts of carbon monoxide, they may be sent via
gas supply pipe 64 to an optional CO boiler 65 in which the CO is
burned in order to heat heating coil 66 connected with a steam
boiler 67.
The amount of heat passed from the regenerator back to the
riser is preferably controlled, at least in part, by controlling the
flow of catalyst through catalyst s tandpipe 13 by control valve 14
which is preferably a type of slide valve operated by suitable
automatic control equipment (not shown) responsive to the
mperature of product vapors downstream of riser outlet 22. If
additional heat removal from the combustion chambers is required,
conventional direct or indirect cooling techniques may be employed
RI-4078C
~Z~7~9
-64-
in the manner known to persons skilled in the art of designing and
operating regenerators.
Since the preferred conversion reactor is of the progressive
flow type and is operated so that no dense bed of catalyst builds
~; up within the riser, space velocities in the riser are usually high
and will generally fall within the range of about 100 to 600 weight
of hydrocarbon per hour per instantaneous weight of catalyst
occupying the riser volume. In the absence of significant catalyst
build up, the instantaneous catalyst inventory within the riser
10 volume if represented by the catalyst particles suspended with the
oil feed.
A particularly preferred embodiment if described in Figure 8
where reference numeral 80 identifies a feed control valve in
feedstock supply pipe 82. Supply pipe 83 (when used) introduces
15 liquid water into the feed. Heat exchange 81 in supply pipe 82
acts as a feed preheater, whereby preheated feed material may be
delivered to the bottom of riser type reactor 91. Catalyst is
delivered to the reactor through catalyst standpipe 86, the flow of
catalyst being regulated by a control valve 87 and suitable
20 automatic control equipment (not shown) with which persons skilled
in the art of designing and operating riser type cracking units are
familiar.
The reactor is equipped with a disengagement chamber 92
similar to the disengagement chamber 19 of the reactor shown in
25 Figure 7. Catalyst departs disengagement chamber 92 through
stripper 94. Spent catalyst passes from stripper 94 to regenerator
101 via spent catalyst transfer pipe 97 having a slide valve 98 for
controlling flow.
RI-~078C
7~
-65-
Regenerator 101 is divided into upper chamber 102 and lower
chamber 103 by a divider panel 104 intermediate the upper and
lower ends of the regenerator. The spent catalyst from transfer
pipe 97 en~ers upper chamber 102 in which the catalyst is partially
5 regeneratad. A funnel-like collector 106 having a bias-cut upper
edge reoeive partially regenerated catalyst from the upper surface
of the dense phase of catalyst in upper chamber 102 and delivers it
via drop leg 107 having an outlet 110 beneath the upper surface of
the dense phase of catalyst in lower chamber 103. Instead of
1() internal catalyst drop leg 107, one may use an external drop leg.
Valve means in such external drop leg can control the residence
time and flow rate in and between the upper and lower chambers.
Air is supplied to the regenerator through an air supply pipe
113. A portion of the air travels through a branch supply pipe 114
to bayonet 115 extending upwardly in the interior of plenum 111
along its central axis. Catalyst in charnber 103 has access to the
space within plenum 111 between its walls and bayonet 115. A small
bayonet (not shown) in the aforementioned space fluffs the catalyst
and urges it upwardly toward a horizontally arranged ring
20 distributor (not shown) where the open top of plenum 111 opens
into chamber 103. The remainder of the air passing through air
supply pipe 113 may be heated in air heater 117 (at least during
start-up with VGO) and is then introduced into inlet 118 of the
ring distributor, which may be provided with holes, nozzles or
25 other apertures which produce an upward flow of gas to fluidize the
partially regenerated catalyst in chamber 103.
The air in chamber 103 completes the regeneration of the
partially regenerated catalyst received via drop leg 107. The
amount of air supplied so that the resultant combustion gases are
still able to support combustion upon reaching top of chamber 103.
Drop lag 107 extends through an enlarged aperture in panel 104, to
RI-4078C
2~ g
-66-
which is secured a gas distributor 120 which is concentric with and
surrounds the drop leg. Combustion supporting gases, which have
been partially depleted, are introduced via gas distributor 120 into
upper regenerator chamber 102 where they contact incoming coked
5 catalyst from coked catalyst transfer pipe 97. Apertured probes
121 in gas distributor 120 assist in achieving a uniform distribution
of the partially depleted combus-tion supporting gas into upper
chamber 102. Supplemental air or cooling fluids may be introduced
into upper chamber 102 -through a supply pipe 122, which may
10 discharge through gas distributor 120.
Fully regenerated catalyst with less than about 0. 25% carbon,
preferably less than about 0.1% and more preferably less than about
0 . 05%, is discharged from lower regenerator chamber 103 through
regenerated catalyst stripper 128, whose outlet feeds into catalyst
15 standpipe 86. Thus, regenerated catalyst is returned to riser 91
for contact with additional fresh feed.
The division of the regenerator into upper and lower
regeneration chambers 102 and 103 not only smooths out variations
in catalyst regenerator residence time but is also uniquely of
assistance in restricting the quantity of regeneration heat which is
imparted to the fresh feed while yielding a regeneration catalyst
with low levels of coke and/or carbon black for return to the riser.
Because of the arrangement of the regenerator, coked catalyst
from transfer line 97 and/or virgin carbon black càtalyst from
addition line 99, with a high loading of carbon, contacts in chamber
102 combustion supporting gases which have already been at least
partially depleted of oxygen by the burning of carbon from partially
regenerated catalyst in lower chamber 102. Because of this, it is
possible -to control both the combustion of carbon and the quantity
30 of carbon dioxide produced in upper regenerator chamber 102.
RI -4078C
~ILZ~47~9
-67--
although regenerating gas introduced through air supply pipe 113
and branch conduit 114 may contain relatively large quantities of
oxygen, the partially regenerated catalyst which it contacts in lower
chamber 103 has already had a major portion of its carbon removed.
S The high oxygen concentration and temperature in chamber 103
combine to rapidly remove the remaining carbon in the catalyst,
thereby achieving a clean, regenerated catalyst wi-th a minimum of
heat release. Thus, here again, the combustion temperature and
CO to C02 ratio in the lower chamber are readily controlled. The
regeneration off gases are discharged from upper chamber 102 via
gas pipe 123, regulator valve 124, catalyst fines trap 125 and outlet
126.
The vapor products from disengagement chamber 92 may be
processed in any convenien t manner such as by discharge through
vapor line 131 to fractionator 132. Fractionator 132 includes a
bottoms outlet 133, side outlet 134, flush oil stripper 135, and
stripper bottom line 136 connected to pump 137 for discharging
flush oil. Overhead product from stripper 135 returns to
fractionator 132 via line 138.
The main overhead discharge line 139 of the fractionator is
connected to receiver 142 having a bottoms line 143 feeding into
pump 144 for discharging gasoline product. A portion of this
product may be returned to the fractionator via recircula~on line
145, the flow being controlled by valve 146. The receiver also
includes a water receiver 147 and a water discharge line 148. The
gas outlet 150 of the overhead receiver discharges a stream which
is mainly below C5, but containing some C5, C6 and C7 material.
If desired, the C5 and above material in the gas stream may be
separated by compression, cooling and fractionation, and recycled
to receiver 142.
RI -4078C
~244~7~39
-68-
The oxidizing gas, such as air, introduced into regeneration
zone 103 through line 114 may be mixed with a cooling spray of
water from a conduit 109. The mixture of oxidizing gas and
atomized water flows through bayonet 115 and thus into the lower
5 bed of catalyst particles.
-
The apertures in distributor 120 are large enough so that theupwardly flowing gas readily passes therethrough into zone 102.
However, the perforations are sized so that the pressure difference
between the upper and lower zones prevents catalyst partic]es from
10 passing downwardly through the distributor. The bayonet 115 and
distributor are simi~arly sized. Gases exiting the regenerator
comprise combustion products, nitrogen, steam formed by
combustion reactions and/or from vaporizing water added with the
oxidizing gas, and oxides of sulfur and other trace elements.
15 These gases are separated from suspended catalyst particles by a
cyclone separator (not shown) and then pass out of the regenerator
through discharge conduit 123.
While this invention may be used with single stage
regenerators, or with the multiple stage regenerator of Figure 7,
20 which has basically concurrent instead of countercurrent flow, it is
especially useful in a regenerator of the type shown in Figure 8,
which is well-suited for producing combustion product gases having
a high ratio of CO to C02, which helps lower regeneration
temperatures in the presence of high carbon levels.
THE CONVERSION PROCESS
The catalysts described in this specification may be employed
in the processes and apparatuses for carbo-metallic oil conversion
described in our U.S. Patents 4,299,687; 4,332,673; 9,341,624;
4,397,122 and 4,354,923.
Rl -4078C
~ . .
~24~7~3~
- 69 -
Furthermore, the adverse effects of vanadium deposited
on the catalyst may be controlled as desaribed in co-pending
international application no. PCT/SU81/00356 of ~shland Oil,
Inc., filed on March 19, 1981, and entitled Immobilization of
Vanadia Deposited on Catalytic Materials During Carbo-metallic
Oil Conversion.
A prefsrred method for cracking feeds with heavy
fractions, referred to as Reduced Crude Conversion (RCC) after
a particularly common and useful carbo-metallic feed, is
disclosed in U.S. Patent 4,341,624. The preferred feeds
capable of being cracked by this method are comprised of 100
percent or less of the 343C+ (650FI) material previously
described. The cracking reaction according to the method
disclosed in U.S. Patent 4,341,624 is sufficiently severe to
convert 50 to 90 percent of the carbo-metallic oil feed to
gasoline per pass and produce coke in amounts of 4 to 14
percent by weight based on weight of fresh feed. This coke is
laid down on the catalyst in amounts in the range of about 0.3
to 3 percent by weight of catalyst, depending upon the catalyst
to oil ratio (weight of catalyst to weight of feedstock) in the
riser.
The feed, with or without pretreatment, is introduced as
shown in Figure 7 into the bottom of the riser along with a
suspension of hot cracking catalyst prepared in accordance with
this invention. Steam, naphtha, water and/or some other
diluent is preferably introduced into the riser along with the
feed. These additives may be from a fresh source or may be
recvcled from a process stream in the refinery. Where recycle
additive streams are.
-70-
used, they may contain hydrogen sulfide and other sulfur
compounds which may help passivate adverse catalytic activity by
heavy metals accumulating on the catalyst. It is to be understood
that the water additive may be introduced either as a li~uid or as
5 steam. Water is added primarily as a source of vapor for
accelerating the feed and catalyst to achieve the vapor velocity and
residence times desired. Other diluents as such need not be added
but where used, the total amount of diluent specified includes the
amount of water used. Extra diluent would further increase the
:10 vapor velocity and further lower the feed partial pressure in the
riser .
As the feed travels up the riser, it is catalytically cracked to
form basically five products known in the industry as dry gas, wet
gas, cat naphtha, light cycle oil, and heavy cycle and/or slurry
15 oil. A-t the upper end of the riser, the catalyst particles are
ballistically separated from product vapors as previously described.
The catalyst which then contains the coke formed in the riser is
sent to the regenerator to burn off the coke and the separated
product vapors are sent to a fractionator for further separation and
20 treatment to provide the five basic product indicated. The
preferred process conditions for the riser conversion reaction are
set forth in Table 3, in which the abbreviations used have the
following meanings: "Temp. " for temperature, "Dil. " for diluent,
"pp" for partial pressure, "wt" for weight, "V" for vapor, "Res. "
25 for residence, "C/O" for catalyst to oil ratio, "Cat. " for catalyst,
"bbl" for barrel, "MAT" for microactivity by the MAT test using a
standard Davison feedstock, "Vel. " for velocity, "cge" for charge,
"d" for density, and "Reg." for regenerated.
RI -4078C
~2~7~9
-71 -
Table 6 - RCC Riser Conditions
Board
Operatins~ Preferred
Parameter Range Range
Feed Temp. 204-427C 204-338C
(400-800F) (400-640~F)
Steam Temp. 93-260C 149-204C
(200-500F) (300-400F)
Reg. Catalyst Temp. 593-815C 690-787C
(1100-1500F) (1275-1450F)
Riser Exit Temp.482-760C 510-593C
(900-1400F) (950-1100F)
Pressure 0-100 psia 10-50 psia
Water/Feed 0.05-0.3Q 0.05-0.15
Dil. pp/Feed pp0.25-3.0 1.0-2.5
Dil . wt/Feed wt_0.4 0.1-0.3
V . Res . Time 0.1-5 0.5-3 sec .
C/O, wt. 3-18 5-12
Lbs . Cat/bbl Feed 0.1-4.0 0.2-2.0
Inlet Cat. MAT>50 vol% >60
Outlet Cat. MAT>20 vol% _40
V. Vel. 25-90 ft/sec 30-60
V . Vel . /Cat . Vel . >1.0 1.2-2.0
Dil. Cge. Vel.5-90 ft/sec 10-50
Oil Cge. Vel 1-50 ft/sec 5-50
Inlet cat. d.1-9 lbs/ft3 2-6
Outlet cat. d.1-6 Ibs/ft3 1-3
Coke, wt% coked cat. 0.6-1.5 0.8-1.3
RI-4078C
-72-
2~7~9
In crackina carbo-metallic feedstocks in accordance with the
method of ~3.S Patent 4,341,624, the-regenerat~ng gas may be any aas which
can provide oxygen to convert carbon to carbon oxides. Air is
highly suitable for this purpose in view of its ready availability.
5 The amount of air required per pound of coke for combustion
depends upon the desired carbon dioxide to carbon monoxide ratio
in the effluent gases and upon the amount of other combustible
materials present in the coke, such as hydrogen, sulfur, ni trogen
and other elements capable of forming gaseous oxides at regenerator
10 conditions.
The regenerator is operated at temperatures in the range of
about 538 to 871C (1000 to 1600F), preferably 690 to 787C (].275
to 1450F), to achieve adequate combustion while keeping catalyst
temperatures below those at which significant catalyst degradation
15 can occur. In order to control these temperatures, it is necessary
to control the rate of burning which in turn can be controlled at
least in part by the relative amounts of oxidizing gas and carbon
introduced into the regeneration zone per unit time. The rate of
introducing carbon into the regenerator may be controlled by
~0 regulating the rate of flow of coked catalyst through valve 40 in
conduit 3~, the rate of removal of regenerated catalyst by
regulating valve 14 in conduit 13, and the rate of introducing
oxidizing gas by the speed of operation of blowers 41, 42. These
parameters are regulated such that the ratio of carbon dioxide to
25 carbon monoxide in the effluent gases is equal to or less than abou~
4.Q, preferably 1.5 or less. In addition, water, either as liquid or
steam, may be added to the regenerator to help control
temperatures and to influence the carbon dioxide to carbon
monoxide ratio.
The regenerator combustion reaction is carried out so that the
amount of carbon remaining on regenerated catalyst is less than
RI -4078C
89
-73-
about 0 . 25, preferably less than about 0 . 05 percent and most
preferably less than about 0 . 01 percent on a substantially
moisture-free weight basis. The residual carbon level is
ascertained by conventional techniques which include drying the
catalyst at 593C (1100F) for about four hours before measuring
carbon conten-t.
In the rnethod of regeneration of the present invention, the
amount of oxidizing gas and catalyst are controlled so that the
amount of oxidizing gas passing into the second zone is greater
ln than the required to convert all of the coke remaining on the
catalyst reaching this oxygen rich zone to carbon dioxide. On -the
other hand, the amount of oxidizing gas passing into the first zone
from the second zone and the additional oxidizing gas added to the
first zone, such as through line 116 and distributor 118 in Figure
8, is insufficient to convert all of the coke in this zone to carbon
dioxide. As the first zone is therefore oxygen deficient, significant
afterburning and excessive temperatures in the first zone are
prevented, thereby protecting both the catalyst and regeneration
equipment from damage.
Industrial Applicability
The present invention is particularly useful in the catalytic
cracking of high boiling carbo-metallic feedstocks to lower boiling
hydrocarbon fractions in the liquid fuel range. Examples of these
oils are reduced crudes and other crude oils or crude oil frac-tions
25 containing residua as hereinabove defined. However, it is to be
understood that the catalyst and processes of the present invention
are useful for cracking almost any crude oil or crude oil fraction,
including conventional light and heavy gas oils.
~I-4078C
~LZ~7~9
-74-
Although the process is preferably conducted in a riser
reactor of the vented type, other types of risers with either
upward or downward flow or non-riser type reactors may be
employed. Thus, the cracking operation may be conducted with a
S fluid bed of catalyst through which the feedstock is passed under
suitable contact conditions of pressure, temperature and feed rate.
Alternately, the catalyst may be used in the ~orm of a moving bed
passing through or otherwise contacted with the feedstock materials
to be cracked.
Although the preferred contacting opera-tion is catalytic
cracking, the catalyst and processes of the invention may be
employed in various other types of hydrocarbon conversion
operations, such as dehydrocyclization, hydrocracking,
hydroforming of naphthene hydrocarbons and the like,
15 polymerization of olefins, depolymerization of polymers, alkylation,
dealkylation, disproportionation, reforming of naphthas,
isomerization of paraffins and the like, aromatization of paraffins
and the like, hydrogenation, dehydrogenation, various types of
hydrofining operations in which one or more characteristics of the
20 feedstock are improved by treatment with hydrogen in the presence
of a catalyst, ox~dation of organic compounds with an oxidation
media such as air, adsorption and absorption operations, and like
types of other contacting, conversion and/or separation processes.
With the present invention, greater cracking selectivity to
25 gasoline boiling range products and less carbon make for a given
level of conversion may be achieved because of the greater
accessibility of reaction sites to reactants and reduction in
undesirable side reactions. Also the processes may be operated at
higher space velocities and~or lower reaction temperatures to
3Q achieve a given selectivity, conversion and octane number as
compared -to conventional processes employing conventional
catalys~s. The catalyst is generally easier -to regenerate than prior
RI-4073C
~2~
-75 -
art catalyst whether by conventional regeneration processes or
those of the invention because diffusion limitations on oxidation
reactions within the catalyst are minimized.
The large pore CB catalyst has high tolerances to both metals
S and coke and are therefore highly suitable for cracking any charge
s-tock containing metal contaminates and/or coke precursors. By
"high tolerance" is meant that the catalyst can accumulate
significantly greater quantities of poison metals and coke as
compared to prior art catalysts and still have effective catalytic
10 activity in the processes of the invention. During the conversion
reaction, metals and coke may deposit on the walls of the large
feeder pores, rather than on the zeolite where they can block
substantial numbers of catalytic sites. The large feeder pore
matrix thereby prolongs the useful life of the zeolite promoter.
15 The large feeder pores are also capable of absorbing asphaltenes
and other liquid hydrocarbons boiling above reactor temperatures
without becoming filled to a degree significantly retarding diffusion
of lower boiling reactant molecules.
Because the large feeder pores provide better diffusion of
20 large molecules they may be cracked at acidic sites instead of
thermally at or near the catalyst surface. There is therefore less
coke laydown for a given level of conversion. Diffusion during
regeneration is also improved so -that temperatures of localized hot
spots will be considerably less and correspondingly there is less
25 matrix and zeolite pore collapse due to sintering. Furthermore,
with larger pore diameters, partial collapse of feeder pores restrict
hydrocarbon diffusion to a lesser degree than with conventional
catalysts. The large feeder pores are also less likely to be blocked
by the disposition of heavy metals and coke on particle surfaces.
RI-4078C
` ~L;2~47~9
-76-
One factor reducing the quantities of coke to be removed in
the regenerator is the relative ease with which both uncracked and
cracked hydrocarbons can be stripped from the large feeder pores.
Another factor reducing coke deposits is that asphaltenes and other
5 hydrocarbons remaining in a liquid state are more likely to be
cracked by reaching an acid site either in the matrix material or on
an exposed surface of the zeoli-te.
A further industrial advantage of the invention is the ease
with which a gel slurry of the catalyst composite can be spray
10 dried and formed into attrition resistant microspheres of a
controlled, flu;dizable size. After formation, these microspheres
may be easily washed and redried and then fired -to remove the
carbon black wi thout significant alteration of either the zeolite or
the support structure of the matrix. An especially impor-tant
15 feature of the invention is that the carbon black can be removed
during initial circulation of the catalyst through the reactor and
regenerator of an operating conversion unit, rather than requiring
removal during catalyst manufacture.
RI-4078C
